Skip to main content
Internet Archive's 25th Anniversary Logo

Full text of "Growth and structure of the English language"

See other formats



;rowth and structure 
)f the english language 


OTTO JESPERSEN, ph.d., lit.d., 


"lehrbuch der phonetik", "phonetische grundfragen", 
"how to teach a foreign language", 
"a modern english grammar", etc. 






"^I^ -10-76 


The scope and plan of this volume have been set 
forth in the introductory paragraph. I have endeavoured 
to write at once popularly and so as to be of some 
profit to the expert philologist. In some cases I have 
advanced new views without having space enough to 
give all my reasons for deviating from commonly 
accepted theories, but I hope to find an opportunity 
in future works of a more learned character to argue 
out the most debatable points. 

I owe more than I can say to numerous predecessors 
in the fields of my investigations, most of all to the 
authors of the New English Dictionary. The dates given 
for the first and last appearance of a word are nearly 
always taken from that splendid monument of English 
scholarship, and it is hardly necessary to warn the 
reader not to take these dates too literally. When I say, 
for instance, th^it fenester was in use from 1 290 to 1548, 
I do not mean to say that the word was actually heard 
for the first and for the last time in those two years, 
but only that no earlier or later quotations have been 
discovered by the painstaking authors of that dictionary. 

I have departed from a common practice in retaining 
the spelling of all authors quoted. I see no reason why 
in so many English editions of Shakespeare the spelling is 
modernized while in quotations from other Elizabethan 
authors the old spelling is followed. Quotations from 
Shakespeare are here regularly given in the spelling of 
the First FoHo (1623). The only point where, for the 
convenience of modern readers, I regulate the old usage, 


IV Preface. 

is with regard to capital letters and «, z', z*, /, printing, 
for instance, us and love instead of vs and loue. — To 
avoid misunderstandings, I must here expressly state 
that by Old English (O. E.) I always understand the 
language before 1150, still often termed Anglo-Saxon. 
I want to thank Mr. A. E. Hayes of London, Dr. 
Lane Cooper of Cornell University, and especially Pro- 
fessor G. C. Moore Smith of Sheffield University, who 
has in many ways given me the benefit of his great 
knowledge of the English language and of English 

In the second edition I have here and there modi- 
fied an expression, added a fresh illustration, and 
removed a remark or an example that was not per- 
haps very felicitously chosen; but in the main the 
work remains unchanged. 

Gentofte (Copenhagen), September 191 1. 



Chapter I ^*^® 

Preliminary Sketch ^ 

Chapter II 
The Beginnings ^^ 

Chapter III 
Old English 33 

Chapter IV 
The Scandinavians 59 

Chapter V 
The French 84 

Chapter VI 
Latin and Greek iH 

Chapter VII 
Various Sources 152 

Chapter VIII 
Grammar ^7^ 

Chapter IX 
Shakespeare and the Language of Poetry . . 210 

Chapter X 
Conclusion 234 

Phonetic Symbols. Abbreviations . . . 249 
Index 250 

Chapter /. 

Preliminary Sketch. 

1. It will be my endeavour in this volume to character- 
ize the chief peculiarities of the English language, and 
to explain the growth and significance of those features 
in its structure which have been of permanent import- 
ance. The older stages of the language, interesting as 
their study is, will be considered only in so far as they 
throw light either directly or by way of contrast on the 
main characteristics of present-day English, and an at- 
tempt will be made to connect the teachings of linguistic 
history with the chief events in the general history of the 
English people so as to show their mutual bearings on 
each other and the relation of language to national 
character. The knowledge that the latter conception is a 
very difficult one to deal with scientifically, as it may 
easily tempt one into hasty generalizations, should make 
us wary, but not deter us from grappHng with problems 
which are really both interesting and important. — My 
plan will be, first to give a rapid sketch of the language 
of our own days, so as to show how it strikes a foreigner 
— a foreigner who has devoted much time to the study 
of English, but who feels that in spite of all his efforts 
he is only able to look at it as a foreigner does, and not 
exactly as a native would — and then in the following 
chapters to enter more deeply into the history of the 
language in order to describe its first shape, to trace the 

Jespershn: English. 2nd ed. I 

,2%; Ic J V; ' ' ' I.'- Pvelim'inary Sketch. 

various foreign influences it has undergone, and to give 
an account ot its own inner growth. 

2, It is, of course, impossible to characterize a lan- 
guage in one formula; languages, like men, are too com- 
posite to have their whole essence summed up in one 
short expression. Nevertheless, there is one expression 
that continually comes to my mind whenever I think 
of the English language and compare it with others: it 
seems to me positively and expressly masculine, it is 
the language of a grown-up man and has very little 
childish or feminine about it. A great many things go 
together to produce and to confirm that impression, 
things phonetical, grammatical, and lexical, words and 
turns that are found, and words and turns that are not 
found, in the language. In dealing with the English 
language one is often reminded of the characteristic 
English hand-writing; just as an English lady will nearly 
always write in a manner that in any other country 
would only be found in a man's hand, in the same manner 
the language is more manly than any other language I 

3. First I shall mention the sound system. The English 
consonants are well defined; voiced and voiceless con- 
sonants stand over against each other in neat symmetry, 
and they are, as a rule, clearly and precisely pronounced. 
You have none of those indistinct or half-slurred con- 
sonants that abound in Danish, for instance (such as 
those in ha^e, hage, liz;lig) where you hardly know 
whether it is a consonant or a vowel-glide that meets 
the ear. The only thing that might be compared to this 
in English, is the r when not followed by a vowel, but 
then this has really given up definitely all pretensions 
to the rank of a consonant, and is (in the pronunciation 
of the South of England) either frankly a vowel (as in 
here) or else nothing at all (in hart, etc.). Each English 

Sound System. ^ 

consonant belongs distinctly to its own type, a / is a /, 
and a ^ is a ^, and there an end. There is much less 
modification of a consonant by the surroundmg vowels 
than in some other languages, thus none of that palatal- 
ization of consonants which gives an insinuating grace 
to such languages as Russian. The vowel sounds, too, 
are comparatively independent of their surroundings, and 
in this respect the language now has deviated widely 
from the character of Old English and has become more 
clear-cut and distinct in its phonetic structure, although, 
to be sure, the diphthongization of most long vowels 
(in ale, whole, eel, who, phonetically eil, houl, ijl, huw) 
counteracts in some degree this impression of neatness 
and evenness. 

4. Besides these characteristics, the full nature of 
which cannot, perhaps, be made intelligible to any but 
those familiar with phonetic research, but which are still 
felt more or less instinctively by everybody hearing the 
language spoken, there are other traits whose importance 
'can with greater ease be made evident to anybody 
possessed of a normal ear. 

5. To bring out cleaily one of these points I select at 
random, by way of contrast, a passage from the language 
of Hawaii: 'T kona hiki ana aku ilaila ua hookipa ia 
mai la oia me ke aloha pumehana loa.'' Thus it goes 
on, no single word ends in a consonant, and a group of 
two or more consonants is never found. Can any one 
be in doubt that even if such a language sound plea- 
santly and be full of music and harmony, the total im- 
pression is childlike and effeminate? You do not expect 
much vigour or energy in a people speaking ouch a lan- 
guage; it seems adapted only to inhabitants of sunny 
regions where the soil requires scarcely any labour on 
the part of man to yield him everthing he wants, and 
where life therefore does not bear the stamp of a hard 


A I. Preliminary Sketch, 

struggle against nature and against fellow-creatures. In 
a lesser degree we find the same phonetic structure in 
such languages as Italian and Spanish; but how different 
are our Germanic tongues. English has no lack of words 
ending in two or more consonants, — I am speaking, 
of course, of the pronunciation, not of the spelling — 
age, hence, wealth, tent, tempt, tempts, months, helped, 
feasts, etc. etc., and thus requires, as well as presupposes, 
no little energy on the part of the speakers. That many 
suchlike consonant groups do not tend to render the 
language beautiful, one is bound readily to concede; 
however, it cannot be pretended that their number in 
English is great enough to make the language harsh or 
rough. While the fifteenth century greatly increased the 
number of consonant groups by making the e mute in 
monthes, helped, etc., the following centuries, on the con- 
trary, lightened such groups as -ght in night, thought 
(where the "back-open'' consonant as German ch is still 
spoken in Scotch) and the initial kn-, gn- in know, 
gnaw, etc. Note also the disappearance of / in alms,' 
folk, etc., and of r in hard, court, etc.; the final conso- 
nant groups have also been simplified in comb and the 
other words in -mb (whereas b has been retained in 
timber) and in the exactly parallel group -ng, for in- 
stance in strong, where now only one consonant is heard 
after the vowel, a consonant partaking of the nature of n 
and of g, but identical with neither of them; formerly it 
was followed by a real g, which has been retained in 

6. In the first ten stanzas of Tennyson's "Locksley 
Hall", three hundred syllables, we have only thirty-three 
words ending in two consonants, and two ending in 
three, certainly no excessive number, especially if we 
take into account the nature of the groups, which are 
nearly all of the easiest kind (-dz: comrades, Pleiads; 

Endings. 5 

-mz: gleams, comes; -nz: robin's, man's, turns; -ns: 
distance, science; -ks: overlooks; -ts: gets, thoughts; 
-kts: tracts, cataracts; -zd: reposed, closed; -st: rest, 
West, breast, crest; -Jt: burnish'd; -nd: sound, around, 
moorland, behind, land; -nt: want, casement, went, 
present; -Id: old, world; It: result; -If: himself; -pt: 
dipt). Thus, we may perhaps characterize English, 
phonetically speaking, as possessing male energy, but 
not brutal force. The accentual system points in the 
same direction, as will be seen below (26 — 28). 

7. The Italians have a pointed proverb: "Le parole 
son femmine e i fatti son maschi." If briefness, concise- 
ness and terseness are characteristic of the style of men, 
while women as a rule are not such economizers of 
speech, English is more masculine than most languages. 
We see this in a great many ways. In grammar it has 
got rid of a great many superfluities found in earlier 
English as well as in most cognate languages, reducing 
endings, etc., to the shortest forms possible and often 
doing away with endings altogether. Where German 
has, for instance, alle diejenigen wilden Here, die dort 
leben, so that the plural idea is expressed in each word 
separately (apart, of course, from the adverb), English 
has all the wild animals that live there, where all, the 
article, the adjective, and the relative pronoun are alike 
incapable of receiving any mark of the plural number; 
the sense is expressed with the greatest clearness imagi- 
nable, and all the unstressed endings -e and -en, 
which make most German sentences so drawling, are 

8. Rimes based on correspondence in the last syl- 
lable only of each line (as bet, set; laid, shade) are 
termed male rimes, as opposed to feminine rimes, where 
each line has two corresponding syllables, one strong 
and one weak (as better, setter; lady, shady). It is true 

6 I. Preliminary' Sketch. 

that these names, which originated in France, were not 
at first meant to express any parallelism with the charac- 
teristics of the two sexes, but arose merely from the 
grammatical fact that the weak -e was the ending of 
the feminine gender (grande, etc.). But the designa- 
tions are not entirely devoid of symbolic significance; 
there is really more of abrupt force in a word that ends 
with a strongly stressed syllable, than in a word where 
the maximum of force is followed by a weak ending. 
'Thanks' is harsher and less polite than the two-sylla- 
bled 'thank you'. English has undoubtedly gained in 
force, what it has possibly lost in elegance, by reducing 
so many words of two syllables to monosyllables. If it 
had not been for the great number of long foreign, espe- 
cially Latin, words, English would have approached the 
state of such monosyllabic languages as Chinese, Now 
one of the best Chinese scholars, G. v. d. Gabelentz, 
somewhere remarks that an idea of the condensed power 
of the monosyllabism found in old Chinese may be gath- 
ered from Luther's advice to a preacher 'Geh rasch 'nauf, 
tu's maul auf, hor bald auf.' He might with equal justice 
have reminded us of many English sentences. 'First 
come first served' is much more vigorous than the French 
'premier venu, premier moulu' or 'le premier venu en- 
gr^ne', the German 'wer zuerst kommt mahlt zuerst' 
and especially than the Danish 'den der kommer forst 
til melle, far farst malet'. Compare also 'no cure, no pay', 
'haste makes waste, and waste makes want', 'live and 
learn,* 'Love no man: trust no man: speak ill of no 
man to his face; nor well of any man behind his 
back' (Ben Jonson), 'to meet, to know, to love, and 
then to part' (Coleridge) , 'Then none were for the 
party; Then all were for the state; Then the great 
man help'd the poor. And the poor man loved the 
great' (Macaulay). 


Monosyllabism. 7 

9. It will be noticed, however, — and the quotations 
just given serve to exemplify this, too — that itMs not 
every collocation of words of one syllable that produces 
an effect of strength, for a great many of the short words 
most frequently employed are not stressed at all and 
therefore impress the ear in nearly the same way as pre- 
fixes and suffixes do. There is nothing particularly vigor- 
ous in the following passage from a modern novel: 'It 
was as if one had met part of one's self one had lost for 
a long time', and in fact most people hearing it read 
aloud would fail to notice that it consisted of nothing 
but one-syllable words. Such sentences are not at all 
rare in colloquial prose, and even in poetry they are found 
oftener than in most languages, for instance: — 

And there a while it bode ; and if a man 
Could touch or see it, he was heal'd at once, 
By faith, of all his ills. 

(Tennyson, The Holy Grail:) 

But then, the weakness resulting from many small con- 
necting words is to some extent compensated in Eng- 
lish by the absence of the definite article in a good many 
cases where other languages think it indispensable, e. g. 
•Merry Old England', 'Heaven and Earth'"; 'life is short'; 
'dinner is ready'; 'school is over'; 'I saw him at church', 
and this peculiarity delivers the language from a number 
of those short 'empty words', which when accumulated 
cannot fail to make the style somewhat weak and 

10. Business-like shortness is also seen in such con- 
venient abbreviations of sentences as abound in English, 
for instance, 'While fighting in Germany he was taken 
prisoner' (= while he was fighting). 'He would not 
answer when spoken to.' 'To be left till called for.' 'Once 
at home, he forgot his fears.' 'We had no idea what to 
do.' 'Did they run.> Yes, I made them' (= made them 

8 I. Preliminary Sketch. 

run). 'Shall you play tennis to-day.? Yes, we are going 
to. I should like to, but I can't.' 'Dinner over, he left 
the house.' Such expressions remind one of the abbrevia- 
tions used in telegrams; they are syntactical correspond- 
encies to the morphological shortenings that are also 
of such frequent occurrence in English: cab for cabriolet, 
bus for omnibus, photo for photograph, phone for telephone, 
and innumerable others. 

11. This cannot be separated from a certain sobriety 
in expression. As an Englishman does not like to use 
more words or more syllables than are strictly necessary, 
so he does not like to say more than he can stand to. 
He dislikes strong or hyperbolical expressions of appro- 
val or admiration; 'that isn't half bad' or 'she is rather 
good-looking' are often the highest praises you can draw 
out of him, and they not seldom express the same warmth 
of feeling that makes a Frenchman ejaculate his 'char- 
mant' or 'ravissante' or 'adorable'. German kolossal or 
pyramidal can often be correctly rendered by English 
great or biggish, and where a Frenchman uses his adverbs 
extremement or infiniment, an Englishman says only very 
or rather or pretty. 'Quelle horreur I' is 'That's rather a 
nuisance'. 'Je suis ravi de vous voir* is 'Glad to see you', 
etc. An Englishman does not like to commit himself by 
being too enthusiastic or too distressed, and his language 
accordingly grows sober, too sober perhaps, and even 
barren when the object is to express emotions. There 
is in this trait a curious mixture of something praise- 
worthy, the desire to be strictly true without exaggerat- 
ing anything or promising more than you can perform, 
and on the other hand of something blameworthy, the 
idea that it is affected, or childish and effeminate, to give 
vent to one's feelings, and the fear of appearing ridic- 
ulous by showing strong emotions. But this trait is 
certainly found more frequently in men than in women. 

Sobriety. g 

so I may be allowed to add this feature of the English 
language to the signs of masculinity I have collected. 

12. Those who use many strong words to express their 
likes or dislikes will generally also make an extensive 
use of another linguistic appliance, namely violent changes 
in intonation. Their voices will now suddenly rise to a 
very high pitch and then as suddenly fall to low tones. 
An excessive use of this emotional tonic accent is charac- 
teristic of many savage nations; in Europe it is found 
much more in Italy than in the North. In each nation 
it seems as if it were more employed by women than by 
men. Now, it has often been observed that the English 
speak in a more monotonous way than most other nations, 
so that an extremely slight lising or lowering of the tone 
indicates what in other languages would require a much 
greater interval. 'Les Anglais parlent extr^mement bas', 
says H. Taine [Notes sur V Angleterre, p. 66). 'Une soci^t^ 
italienne, dans laquelle je me suis fourvoy^ par hasard, 
m'a positivement etourdi; je m'6tais habitue k ce ton 
mod^re des voix anglaises.' Even English ladies are in 
this respect more restramed than many men belonging 
to other nations: 

'She had the low voice of your English dames, 
Unused , it seems , to need rise half a note 
To catch attention' 

(Mrs. Browning, Aurora Leigh p. 91).* 

13. If we turn to other provinces of the language we 
shall find our impression strengthened and deepened. 

It is worth observing, for instance, how few diminu- 
tives the language has and how sparingly it uses them. 
English in this respect forms a strong contrast to Italian 
with its -ino (ragazzino, fratellino, originally a double 

I Cf. my Lehrbuch der Phonetik, p. 226; Fonetik (Dan. ed.) 
p. 588. 

lO I. Preliminary Sketch. 

diminutive), -ina (donnina), -etto (giovinetto), -etta 
(oretta), -ello, -ella (asinello, storiella) and other endings, 
German with its -chen und -lein, especially South German 
with its eternal -le, Dutch with its -;>, Russian, Magyar, 
and Basque with their various endings. The continual 
recurrence of these endings without any apparent ne- 
cessity cannot but produce the impression that the 
speakers are innocent, childish, genial beings with no 
great business capacities or seriousness in life. But in 
English there are very few of these fondling-endings; 
•let is in the first place a comparatively modern ending, 
very few of the words in which it is used go back more 
than a hundred years; and then its extensive use in 
modern times is chiefly due to the naturalists who want 
it to express in a short and precise manner certain small 
organs [budlet Darwin ; hladelet Todd ; conelet Dana ; bulb- 
let Gray; leaflet, fruitlet, jeatherlet, etc.) — an employ- 
ment of the diminutive which is as far removed as pos- 
sible from the terms of endearment found in other lan- 
guages. The endings -kin and -ling (princekin, prince- 
ling) are not very frequently used and generally express 
contempt or derision. Then, of course, there is -y, -ie 
(Billy, Dicky, auntie, birdie, etc.) which corresponds 
exactly to the fondling-suffixes of other languages; but 
its application in English is restricted to the nursery and 
it is hardly ever used by grown-up people except in speak- 
ing to children. Besides, this ending is more Scotch 
than English, and the Scotch with all their deadly ear- 
nestness, especially in religious matters, are, perhaps, in 
some respects more childlike than the English. 

14. The business-like, virile qualities of the English 
language also manifest themselves in such things as word- 
order. Words in English do not play at hide-and-seek, 
as they often do in Latin, for instance, or in German, 
where ideas that by right belong together are widely 

Word - order. 1 1 

sundered in obedience to caprice or, more often, to a 
rigorous grammatical rule. In English an auxiliary verb 
does not stand far from its main verb, and a negative 
will be found in the immediate neighbourhood of the 
word it negatives, generally the verb (auxiliary). An 
adjective nearly always stands before its noun; the only 
really important exception is when there are qualifications 
added to it which draw it after the noun so that the 
whole complex serves the purpose of a relative clause: 
*a man every way prosperous and talented' (Tennyson), 
*an interruption too brief and isolated to attract more 
notice' (Stevenson). And the same regularity is found 
in modern English word-order in other respects as well. 
A few years ago I made my pupils calculate statistically 
various points in regard to word-order in different lan- 
guages. I give here only the percentage in some modern 
authors of sentences in which the subject preceded the 
verb and the latter in its turn preceded its object (as 
in 'I saw him' as against 'Him I saw, but not her' or 
'Whom did you see.?'): — 

Shelley, prose 89, poetry 85. 

Byron, prose 93, poetry 81. 

Macaulay, prose 82. 

Carlyle, prose 87. • . 

Tennyson, poetry 88. 

Dickens, prose 91, 

Swinburne, poetry 83. 

Pinero, prose 97. 

For the sake of comparison I mention that one Danish 
prose-writer (J. P. Jacobsen) had 82, a Danish poet 
(Drachmann) 61, Goethe (poetry) 30, a modern German 
prose-writer (Tovote) 31, Anatole France 66, Gabriele 
d'Annunzio 49 per cent of the same word-order. That 
English has not always had the same regularity, is shown 
by the figure for Beowulf being 16, and for King Alfred's 

12 I. Preliminary Sketch. 

prose 40. Even if I concede that our statistics did not 
embrace a sufficient number of extracts to give fully 
reliable results, still it is indisputable that English shows 
more regularity and less caprice in this respect than most 
or probably all cognate languages, without however, 
attaining the rigidity found in Chinese, where the per- 
centage in question would be lOO (or very near it). Eng- 
lish has not deprived itself of the expedient of inverting 
the ordinary order of the members of a sentence when 
emphasis requires it, but it makes a more sparing use 
of it than German and the Scandinavian languages, and 
in most cases it will be found that these languages 
emphasize without any real necessity, especially in a 
great many every-day phrases: 'daer har jeg ikke vaeret', 
'dort bin ich nicht gewesen', 'I haven't been there'; 
'det kan jeg ikke', *das kann ich nicht', 'I can't do that'. 
How superfluous the emphasis is, is best shown by the 
usual phrase, 'det veed jeg ikke', 'das weiss ich nicht', 
where the Englishman does not even find it necessary 
to state the object at all: 'I don't know.' Note also that 
in English the subject precedes the verb after most intro- 
ductory adverbs: 'now he comes'; 'there he goes', while 
German and Danish have, and English had till a few 
centuries ago, the inverted order: 'jetzt kommt er', 'da 
geht sie' ; 'nu kommer han', 'daer gar hun' ; 'now comes he', 
'there goes she'. Thus order and consistency signalize 
^ the modern stage of the English language. 

15. No language is logical in every respect, and we 
must not expect usage to be guided always by strictly 
logical principles. It was a frequent error with the older- 
grammarians that whenever the actual grammar of a 
language did not seem conformable to the rules of ab- 
stract logic they blamed the language and wanted to 
correct it. Without falling into that error we may, never- 
theless, compare different languages and judge them by 

Logic. 13 

the standard of logic, and here again I think that, apart 
from Chinese, which has been described as pure applied 
logic, there is perhaps no language in the civilized world 
that stands so high as English. Look at the use of the 
tenses; the difference between the past he saw and the 
composite perfect he has seen is maintained with great 
consistency as compared with the similarly formed tenses 
in Danish, not to speak of German, so that one of the 
most constant faults committed by English-speaking 
Germans is the wrong use of these forms ('Were you in 
Berlin?' for 'Have you been in (or to) Berlin?', 'In 181 5 
Napoleon has been defeated at Waterloo' for 'was de- 
feated'). And then the comparatively recent develop- 
ment of the extended (or 'progressive') tenses has furnished 
the language with the wonderfully precise and logi- 
cally valuable distinction between 'I write' and 'I am 
writing', 'I wrote' and 'I was writing'. French has 
something similar in the distinction between le pass6 
defini (j'ecrivis) and I'imparfait (j'6crivais), but on the 
one hand the former tends to disappear, or rather has 
already disappeared in the spoken language, at any rate 
in Paris and in the northern part of the country, so that 
fai ecrit takes its place and the distinction between 
'I wrote' and 'I have written' is abandoned; on the other 
hand the distinction applies only to the past while in 
English it is carried through all tenses. Furthermore, 
the distinction as made in English is superior to the 
similar one found in the Slavonic languages, in that it is 
made uniformly in all verbs and in all tenses by means 
of the same device [am -ing), while the Slavonic languages 
employ a much more complicated system of prepositions 
and derivative endings, which has almost to be learned 
separately for each new verb or group of verbs. 

16. In praising the logic of the English language we 
must not lose sight of the fact that in most cases where, 

I A I. Preliminary Sketch. 

so to speak, the logic of facts or of the exterior world 
is at war with the logic of grammar, English is free from the 
narrow-minded pedantry which in most languages sacrifices 
the former to the latter or makes people shy of saying 
or writing things which are not 'strictly grammatical'. 
This is particularly clear with regard to number. Family 
and clergy are, grammatically speaking, of the singular 
number; but in reality they indicate a plurality. Most 
languages can treat such words only as singulars, but inr 
English one is free to add a verb in the singular if the 
idea of unity is essential, and then to refer to this unit 
as it, or else to put the verb in the plural and use the 
pronoun they, if the idea of plurality is predominant. It 
is clear that this liberty of choice is often greatly advan- 
tageous. Thus we find sentences like these, 'As the clergy 
are or are not what they ought to be, so are the rest of 
the nation' (Miss Austen), or 'the whole race of man 
(sing.) proclaim it lawful to drink wine' (De Quincey), 
or 'the club all know that he is a disappointed man' (the 
same). In 'there are no end of people here that I don't 
know' (George Eliot) no end takes the verb in the plural 
because it is equivalent to 'many', and when Shelley 
writes in one of his letters 'the Quarterly are going to 
review me' he is thinking of the Quarterly (Review) as a 
whole staff of writers. Inversely, there is in English a 
freedom paralleled nowhere else of expressing grammati- 
cally a unity consisting of several parts, of saying, for 
instance, 'I do not think I ever spent a more delightful 
three weeks' (Ch. Darwin), 'for a quiet twenty minutes', 
'another United States', cf. also 'a fortnight' (originally 
a fourteen-night) ; 'three years is but short' (Shakespeare), 
'sixpence was offered him' (Ch. Darwin), 'ten minutes 
is heaps of time' (E. F. Benson), etc. etc. 

17. A great many other phenomena in English show 
the same freedom from pedantry, as when passive con- 

Freedom from Pedantry. I e 

structions such as 'he was taken no notice of are allowed, 
or when adverbs or prepositional complexes may be 
used attributively as in 'his then residence/ 'an almost 
reconciliation' (Thackeray), 'men invite their out-College 
friends' (Steadman), 'smoking his before-breakfast pipe' 
(Co. Doyle), 'in his threadbare, out-at-elbow shooting- 
jacket' (G. du Maurier), or when even whole phrases or 
sentences may be turned into a kind of adjective, as in 
'with a quite at home kind of air' (Smedley), 'in the 
pretty diamond-cut-diamond scene between Pallas and 
Ulysses' (Ruskin), 'a little man with a puffy Say-nothing 
to-me-, -or-FU-contradict-you sort of countenance' 
(Dickens), 'With an I-turn-the-crank-of-the-Universe air' 
(Lowell), 'Rose is simply self-willed; a 'she will' or 'she 
won't' sort of httie person' (Meredith). Although such 
combinations as the last-mentioned are only found in 
more or less jocular style, they show the possibilities of 
the language, and some expressions of a similar order be- 
long permanently to the language, for instance, 'a would- 
be artist', 'a stay-at-home man', 'a turn-up collar'. Such 
things — and they might be easily miultiplied — are in- 
conceivable in such a language as French where every- 
thing is condemned that does not conform to a 
definite set of rules laid down by grammarians. 
The French language is like the stiff French garden^ 
of Louis XIV, while the English is like an English 
park, which is laid out seemingly without any definite 
plan, and in which you are allowed to walk 
everywhere according to your own fancy without having 
to fear a stern keeper enforcing rigorous regulations. 
The English language would not have been what it 
is if the English had not been for centuries great 
respecters of the liberties of each individual and if 
everybody had not been free to strike out new paths 
for himself. 

J 5 I- Preliminary Sketch. 

18. This is seen, too, in the vocabulary. In spite of 
the efforts of several authors of high standing, the English 
have never suffered an Acaden^y to be instituted among 
them like the French or Italian Academies, which had 
as one of their chief tasks the regulation of the vocab- 
ulary so that every word not found in their Dictionaries 
was blamed as unworthy of literary use or distinction. 
In England every writer is, and has always been, free 
to take his words where he chooses, whether from the 
ordinary stock of everyday words, from native dialects, 
from old authors, or from other languages, dead or living. 
The consequence has been that English dictionaries com- 
prise a larger number of words than those of any other 
nation, and that they present a variegated picture of 
terms from the four quarters of the globe. Now, it seems 
to be characteristic of the two sexes in their relation to 
language that women move in narrower circles of the 
vocabulary, in which they attain to perfect mastery so 
that the flow of words is always natural and, above all, 
never needs to stop, while men know more words and 
always want to be more precise in choosing the exact 
word with which to render their idea, the consequence 
being often less fluency and more hesitation. It has been 
statistically shown that a comparatively greater number 
of stammerers and stutterers are found among men (boys) 
than among women (girls). Teachers of foreign languages 
have many occasions to admire the ease with which fe- 
male students express themselves in another language 
after so short a time of study that most men would be 
able to say only few words hesitatingly and falteringly, 
but if they are put to the test of translating a difficult 
piece either from or into the foreign language, the men 
will generally prove superior to the women. With regard 
to their native language the same difference is found, 
though it is perhaps not so easy to observe. At any rate 

Vocabulary. j >i 

our assertion is corroborated by the fact observed^^by 
every student of languages that novels written by ladies 
are much easier to read and contain much fewer difficult 
words than those written by men. All this seems to 
justify us in setting down the enormous richness of the 
English vocabulary to the same masculinity of the English 
nation which we have now encountered in so many various 

To sum up: The English language is a methodical, 
energetic, business-like and sober language, that does 
not care much for finery and elegance, but does care for 
logical consistency and is opposed to any attempt to 
narrow-in life by police regulations and strict rules either 
of grammar or of lexicon. As the language is, so also is 
the nation, 

For words, like Nature, half reveal 

And half conceal the Soul within. (Tennyson.) 

Jbspersbn : English, and ed. 

Chapter II. 
The Beginnings. 

20. The existence of the English language as a separ- 
ate idiom began when Germanic tribes had occupied 
all the lowlands of Great Britain and when accordingly 

' the invasions from the continent were discontinued, so 
that the settlers in their ^ew homes were cut off from 
that steady intercourse with their continental relations 
which always is an imperative condition of linguistic 
unity. The historical records of English do not go so far 
back as this, for the oldest written texts in the English 
language (in 'Anglo-Saxon') date from about 700 and are 
thus removed by about three centuries from the begin- j 
nings of the language. And yet comparative philology is 
able to tell us something about the manner in which the 
ancestors of these settlers spoke centuries before that 
period, and to sketch the prehistoric development of what 
was to become the language of King Alfred, of Chaucer \ 
and of Shakespeare. 

21. The dialects spoken by the settlers in England 
belonged to the great Germanic (or Teutonic) branch of 
the most important of all linguistic families, termed by 
many philologists the Indo-European (or Indo-Germanic) 
and by others, and to my mind more appropriately, 
Arian (Aryan). The Arian family comprises a great 
variety of languages, including, besides some languages of 
less importance, Sanskrit with Prakrit and many living 

Primitive Arian. ig 

languages of India; Iranian with Modern Persian; Greek; 
Latin with the modern Romance languages (Italian, 
Spanish, French, etc.); Celtic, two divisions of which still 
survive, one in Welsh and Armorican or Breton, the other 
in the closely connected Irish and Scotch-Gaelic, besides 
the nearly extinct Manx; Baltic (Lithuanian and Lettic) 
and Slavonic (Russian, Czech, Polish, etc.). Among the 
extinct Germanic languages Ulfila's Gothic was the most 
important; the living are High German, Dutch, Low, 
German, Frisian, English, Danish, Swedish, Norwegian, 
and Icelandic. The first five are generally grouped to-' 
gether as West-Germa jjijc^ ^h^s the four last-mentioned 
or Scandinavian languages^onstitute with Gothic the 
East-Germanic group, a grouping which does not, how- 
ever, account for the really much more complex rela- 
tionship between these languages. 

22. The Arian language, which was in course of time 
differentiated into all these languages, or as the same 
, fact is generally expressed in a metaphor of dubious value, 
was the parent-language from which all these languages 
have descended, must by no means be imagined as a 
language characterized by a simple and regular struc- 
ture. On the contrary it must have been, grammatically 
and lexically, extremely complicated and full of irreg- 
ularities. Its grammar was highly inflexional, the 
relations between the ideas being expressed by means 
of endings more intimately fused with the chief element 
of the word than is the case in such agglutinative lan- 
guages as Hungarian (Magyar). Nouns and verbs were 
kept distinct, and where the same sense-modifications 
were expressed in both, such as plurality, it was by means 
of totally different endings. In fact, the indication of 
number — the threefold division into singular, dual, 
and plural — was inseparable from the case-endings in 
the nouns and frorn the person-endings as well as signs 


20 II- The Beginnings. 

of mood and tense in the verbs: one cannot point to 
distinct parts of such a Latin form as est (cantat) or sunt 
(cantant) or fuissem (cantavissem) and say, this element 
means singular (or plural), this one means indicative 
(or subjunctive) and that one indicates what tense the 
whole form belongs to. There were eight cases, but they 
did not, for the greater part, indicate such clear, con- 
crete, outward relations as the Finnic (local) cases do; 
the consequence was a comparatively great number of 
clashings and overlappings, in form as well as in function. 
Each noun belonged to one of three genders, masculine, 
feminine, and neuter; but this division by no means 
corresponded with logical consistency to the natural -j 
division into (l) living beings of one sex, (2) living beings J 
of the other sex, and (3) everything else. Nor did the 
moods and tenses of the verb agree very closely with 
any definite logical categories, the idea of time being, 
moreover, mixed up with that of 'tense-aspect' (in Ger- 
man 'aktionsart'), i. e. distinctions according as an action 
was viewed as momentary or protracted or iterated, etc. 
In the nominal as well as in the verbal inflexions the 
endings varied with the character of the stem they were ' 
added to, and very often the accent was shifted from one 
syllable to another according to seemingly arbitrary 
rules, just as in modern Russian. In a great many cases, 
too, one form was taken from one word and another 
from a totally different one, a phenomenon (called by 
Osthoff 'suppletivwesen') which we have in a few in- 
stances in modern English (good, better; go, went, etc.). 
An idea of the phonetic system of the old Arian language 
may best be gathered from Greek, which has preserved 
the old system with great fidelity on the whole, especially 
the vowels. But of course, no one of the historically 
transmitted languages, not even one of the oldest, can 
give more than an approximate idea of the common Arian 

Germanic, 2 1 

language distant from us by so many thousand years, and 
scholars have now learnt more prudence than was shown 
when Schleicher was bold enough to print a fable in what 
he believed to be a fairly accurate representation of prim- 
itive Arian. 

23. In historical times we find Arian split up into a 
variety of languages, each with its own peculiarities, in 
sounds, in grammar, and in vocabulary. So different were 
these languages that the Greeks had no idea of any 
similarity or relationship between their own tongue and 
that of their Persian enemies; nor did the Romans sus- 
pect that the Gauls and Germans they fought spoke lan- 
guages of the same stock as their own. Whenever the 
Germanic languages are alluded to, it is always in ex- 
pressions like these, 'a Roman tongue can hardly pro- 
nounce such names' or (after giving the names of some 
Germanic tribes) 'the names sound like a noisy war- 
trumpet, and the ferocity of these barbarians adds horror 
even to the words themselves'. Julian the Apostate 
compares the singing of Germanic popular ballads to the 
croaking and shrill screeching of birds. ^ Much of this, 
of course, must be put down to the ordinary Greek and 
Roman contempt for foreigners generally; nor can it 
be wondered at that they did not recognize in these lan- 
guages congeners of their own, for the similarities had 
been considerably blurred by a great many important 
changes in sound and in structure, so that it is only the 
patient research of the nineteenth century that has 
enabled us to identify words in separate languages which 
are now so dissimilar as not to strike the casual observer 
as in any way related. What contributed, perhaps, more 
than anything else to make Germanic words look 
strange, were two great phonetic changes affecting large 

I Kluge, Paul's Grundriss I 354. 

2 2 II- The Beginnings. 

parts of the vocabulary, the consonant- ski ft^ and the 
stress- shift. 

24. The consonant-shift must not be imagined as 
having taken place at one moment; on the contrary it 
must have taken centuries, and modern research has 
begun to point out the various stages in this develop- 
ment. This is not the proper place to deal with detailed 
explanations of this important change, as we must hurry 
on to more modern times; suffice it then to give a few 
examples to show how it affected the whole look of the 
language. Any p was changed to /, — thus we have 
father corresponding to pater and similar forms in the 
cognate languages; any t was made into th [{)], as in 
three, — compare Latin tres\ any k became h, — as 
cornu = horn.^ And as any ^ or ^ or g, any bh, dh, gh 
was similarly shifted, you will understand that there 
were comparatively few words that were not altered past 
recognition; still such there were, for instance mus, now 

I In English books this change ('die erste lautverschiebung ') 
is often called Grimm's law, because the 2nd edition of the first 
volume of Jacob Grimm's Deutsche grammatik (1822) made it 
generally known. But in his first edition (18 19) Grimm did not 
yet know the law; between the two editions he had read the 
Danish scholar Rasmus Rask's Unders^gelse om det gamle nor- 
diske spi'ogs oprindelse (written 18 14, printed 18 18), where the 
sound -correspondences are clearly set forth on p. 169. Grimm 
saw the enormous importance of the discovery and formulated 
the law in a more abstract manner than Rask. As part of the 
law had been seen more or less clearly by a few earlier philo- 
logists, and as Grimm's manner of stating it has been considerably 
modified by recent investigations , the law should not be named 
after any one man. At any rate it is perfectly absurd to extend 
the name of 'Grimm's law' to any similar phonetic change, as 
is sometimes done ('Grimm's Law in South -Africa'). 
L 2 Latin words are here chosen for convenience only as re- 
presenting these old consonants with great fidelity; but of course 
it must not be supposed that the English words named come 
from the Latin. 

Sound Changes. 2 ^ 

mouse, which contained none of the consonants suscep- 
tible of the shifting in question. 

25. The second change affected the general character 
of the language even more thoroughly. Where pre- 
viously the stress was sometimes on the first syllable of 
the word, sometimes on the second, or on the third, etc., 
without any seeming reason and without any regard to 
the intrinsic importance of that syllable, a complete revo- 
lution simplified matters so that the stress rules may be 
stated in a couple of lines: nearly all words were stressed 
on the first syllable; the chief exceptions occurred only 
where the word was a verb beginning with one out of a 
definite number of prefixes, such as those we have in 
modern English beget, forget, overthrow, abide, etc. Verner 
has shown that this shifting of the place of the accent 
took place later than the Germanic consonant - shift, 
and we shall now inquire into the relative importance of 
the two. 

26. The consonant-shift is important to the modern 
philologist, in so far as it is to him the clearest and least 
ambiguous criterion of the Germanic languages: a word 
with a shifted consonant is Germanic, and a word with 
an unshifted consonant in any of the Germanic languages 
must be a loan-word; whereas the shifted stress is no 
such certain criterion, chiefly because many words had 
always had the stress on the first syllable. But if we 
ask about the intrinsic importance of the two changes, 
that is, if we try to look at matters from the point of 
view of the language itself, or rather the speakers, we 
shall see that the second change is really the more im- 
portant one. It does not matter much whether a certain 
number of words begin with a p or with a /, but it does 
matter, or at any rate it may matter, very much whether 
the language has a rational system of accentuation or 
not; and I have no hesitation in saying that the old 

2 4 II. The Beginnings. 

V"' stress-shift has left its indelible mark on the structure 
of the language and has influenced it more than any 
other phonetic change.^ The significance of the stress 
shift will, perhaps, appear most clearly if we compare 
two sets of words in modern English. The original Arian 
I stress system is still found in numerous words taken in 
recent times from the classical languages, thus ^family, 
fa^miliar, famiWarity or ^photograph, phohographer, photo- 
^graphic,^ The shifted Germanic system is shown in such 
groups as Hove, Hover, Hoving, Hovingly, Hovely, Hove- 
liness, Hoveless, Hovelessness, or ^king, ^kingdom, ^kingship, 
^kingly, ^kingless, etc. As it is characteristic of all Arian 
< languages that suffixes play a much greater role than 
prefixes, word-formation being generally by endings, it 
follows that where the Germanic stress system has come 
into force, the syllable that is most important has also 
the strongest stress, and that the relatively insignificant 
modifications of the chief idea which are indicated by 
^ formative syllables are also accentually subordinate. This 
is, accordingly, a perfectly logical system, correspond- 
ing to the piincipal rule observed in sentence stress, viz. 
that the stressed words are generally the most important 
ones. As, moreover, want of stress tends everywhere to 
obscure vowel-sounds, languages with moveable accent 
are exposed to the danger that related words, or different 
forms of the same word, are made more different than 
they would else have been, and their connexion is more 
obscured than is strictly necessary; compare, for in- 
stance, the two sounds in the first syllable of family [seP 

1 Except perhaps the disappearance of so many weak^'s 
about 1400. 

2 I indicate stress by means of a short vertical stroke ' im- 
mediately before the beginning of the strong syllable. 

3 A list of the phonetic symbols used in this book will be 
found on the last page. 



and familiar (9), or the different treatment of the vowels 
in photograph, photographer and photographic. The pho- 
netic clearness inherent in the consistent stress system 
is certainly a linguistic advantage, and the obscuration 
of the connexion between related words is generally to 
be considered a drawback. The language of our fore- 
fathers seems therefore to have gained considerably by 
replacing the movable stress by a fixed one. 

27. The question naturally arises: why was the accent 
shifted in this way.? Two possible answers present them- 
selves. The change may have been either a purely 
mechanical process, by which the first syllable was 
stressed without any regard to signification, or else it 
may have been a psychological process, by which the 
root syllable became stressed because it was the most 
important part of the word. As in the vast majority of 
cases the root syllable is the first, the question must 
be decided from those cases where the two things are 
not identical. Kluge^ infers from the treatment of re- 
duplicated forms of the perfect corresponding to Latin 
cecidi, peperci, etc. that the shifting was a purely mechani- 
cal process; for it was hot the most important syllable 
that was stressed in Gothic haihait 'called', rairo^ 're- 
flected', lailot 'let' (read ai as short e), while in the Old 
English forms of these words heht, reord, leort the vowel 
of the root syllable actually disappears. But it may be 
objected to this view that the reduplicated syllable was 
in some measure the bearer of the root signification, as 
it had enough left of the root to remind the hearer of 
it, and in pronouncing it the speaker had before, him part 
at least of the significant elements. The first syllable of 
a reduplicated perfect must to him have been of a far 
greater importance than one of those prefixes which 

I Paul's Grundriss I 2 389. 

26 II- The Beginnings. 

served only to modify to a small extent the principal 
idea expressed in the root syllable. The fact that 
the reduplicated syllable attracted the accent therefore 
speaks less strongly in favour of the mechanical ex- 
planation than does the want of stress on the verbal 
prefixes in the opposite direction, so that the case 
/seems to me strongest for the psychological theory. 
In other words, we have here a case of value- stressing;'^ 
that part of the word which is of greatest value to 
the speaker and which therefore he especially wants 
the hearer to notice, is pronounced with the strongest 

28. We find the same principle of value-stressing 
everywhere, even in those languages whose traditional 
stress rests or may rest on other syllables than the root 
— this word is here used not in the sense of the ety- 
mologically original part of the word, but in the sense 
of what is to the actual instinct of the speaker intrinsi- 
cally the most significant element — but in these lan- 
guages it only plays the part of causing a deviation from 
the traditional stress now and then whereas in Germanic 
it became habitual to stress the root syllable^, and this 
led to other consequences of some interest. In those 
languages where the stress syllable is not always the most 
significant one, the difference between stressed and un- 
stressed syllables is generally less than in the Germanic 
languages; there is a nicer and subtler play of accent, 
which we may observe in French, perhaps, better than 
elsewhere. In nous chantons the last syllable is stressed, 
but chan- is stronger than for- in Eng. we forget, because 
its psychological value is greater. Where a contrast is 

1 See my Fonetik, Copenh. 1899, p. 557 and 5(^0; Lehrbuch 
def Phonetik, Leipz. 1904, p. 209 ff. 

2 Fonetik, p. 554; Lehrbuch der Phonetik, p. 270. 

Accent. 2 7 

to be expressed it will most often be associated with one 
of the traditionally unstressed syllables, and the result 
is that the contrast is brought vividly before the mind 
with much less force than is necessary in English; in nous 
chantons, et nous ne dansons pas you need not even make 
chan and dan stronger, at any rate not much stronger 
than the endings, while in English we sing, hut we don't 
dance, the syllables sing and dance must be spoken with 
an enormous force, because they are in themselves 
strongly stressed even when no contrast is to be pointed 
out. A still better example is French c'est un acteur et 
non pas un auteur and English he is an actor, and not an 
author; the Frenchman produces the intended effect by 
a slight tap, so to speak, on the two initial syllables of 
the contrasted words, while an Englishman hammers 
or knocks the corresponding syllables into the head of 
the hearer. The French system is more elegant, more 
artistic; the Germanic system is heavier or more clumsy, 
perhaps, in such cases as those just mentioned, but on 
the whole it must be said to be more rational, more logi- 
cal, as an exact correspondence between the inner and 
the outer world is established, if the most significant 
element receives the strongest phonetic expression. This 
Germanic stress-principle has been instrumental in 
bringing about important changes in other respects than 
those considered here. But what has been said here 
seems to me to indicate a certain connexion between 
language and national character; for has it not always 
been considered characteristic of the Germanic peoples 
(English, Scandinavians, Germans) that they say their 
say bluntly without much considering the artistic effect, 
and that they emphasize what is essential without al- 
ways having due regard to nuances or accessory notions.? 
and does not the stress system we have been considering 
present the very same aspect."* 

28 n. The Beginnings. 

29. We do not know in what century the stress was 
shifted^ but the shifting certainly took place centuries 
before the immigration of the English into Great Britain. 
To a similar remote period we must refer several other 
great changes affecting equally all the Germanic lan- 
guages. One of the most important is the simplification 
of the tense system in the verb, no Germanic language 
having more than two tenses, a present and a past. As 
many of the old endings gradually wore off, they were 
not in themselves a sufficiently clear indication of the 
differences of tense, and the gradation (ablaut) of the 
root vowel, which had at first been only an incidental con- 
sequence of differences of accentuation, was felt more 
and more as the real indicator of tense. But neither 
gradation nor the remaining endings were fit to make 
patterns for the formation of tenses in new verbs; con- 
sequently, we see very few additions to the old stock 
of 'strong' verbs, and a new type of verbs, 'weak verbs', 
is constantly gaining ground. Whatever may have been 
the origin of the dental ending used in the past tense of 
these verbs, it is very extensively used in all Germanic 
languages and is, indeed, one of the characteristic fea- 
tures of their inflexional system. It has become the 
'regular' mode of forming the preterite, that is, the one 
resorted to whenever new verbs are called into existence. 

30. To this early period, while the English were still 
living on the Continent with their Germanic brethren, 

I Nothing can be concluded from the existence at the time 
"t)f Tacitus of such series of alliterating names for members of 
the same family as Segestes Segimerus Segimundus, etc. (Kluge, 
Paul's Grundriss ^357, 388) for alliteration does not necessarily 
imply that the syllable has the chief stress of the word; cf. the 
French formulas 7nesse et matines, Florient et Floretie , Basans 
et Basilie, monts et merveilles, quivivraverra, d tortetd travers 
(Nyrop, Grammaire his tongue I *448). 

Loan-words. 2Q 

belong the first class of loan-words. No language is 
entirely pure; we meet with no nation that has not 
adopted some loan-words, so we must suppose that the 
forefathers of the old Germanic tribes adopted words 
from a great many other nations with whom they came 
into contact; and scholars have attempted to point out 
words borrowed very early from various sources. Some 
of these, however, are doubtful, and none of them are 
important enough to arrest our attention before we arrive 
at the period when Latin influence began to be felt in 
the Germanic world, that is, about the beginning of our 
Christian era. But before we look at these borrowings 
in detail, let us first consider for a moment the general 
lesson that may be derived from the study of words 
taken over from one language into another. 

31. Loan-words have been called the milestones of 
philology, because in a great many instances they per- 
mit us to fix approximatively the dates of linguistic 
changes. But they might with just as much right be termed 
some of the milestones of general history, because they 
show us the course of civilization and the wanderings 
of inventions and institutions, and in many cases give 
us valuable information as to the inner life of nations 
when dry annals tell us nothing but the dates of the 
deaths of kings and bishops. When in two languages 
we find no trace of the exchange of loan-words one way 
or the other we are safe to infer that the two nations 
have had nothing to do with each other. But if they 
have been in contact, thef number of the loan-words and 
still more the quality of the loan-words, if rightly inter- 
preted, will inform us of their reciprocal relations, they 
will show us which of them has been the more fertile 
in ideas and on what domains of human activity each 
has been superior to the other. If all other sources of 
information were closed to us except such loan-words 

JO !!• The Beginnings. 

in our modern North- European languages as piano^ 
soprano, opera, libretto, tempo, adagio, etc., we should still 
have no hesitation in drawing the conclusion that Italian 
music has played a great role all over Europe. Similar 
instances might easily be multiplied, and in many ways 
the study of language brings home to us the fact that 
when a nation produces something that its neighbours 
think worthy of imitation these will take over not only 
the thing but also the name. This will be the general 
rule, though exceptions may occur, especially when a 
language possesses a native word that will lend itself 
without any special effort to the new thing imported 
from abroad. But if a native word is not ready to hand 
it is easier to adopt the ready-made word used in the 
other country, nay this foreign word is very often im- 
ported even in cases where it would seem to offer no 
great difficulty to coin an adequate expression by means 
of native word-material. As, on the other hand, there 
is generally nothing to induce one to use words from 
foreign languages for things one has just as well at home, 
loan-words are nearly always technical words belonging 
to one special branch of knowledge or industry, and may 
be grouped so as to show what each nation has learnt 
from each of the others. It will be my object to go 
through the different strata of loans in English with 
special regard to their significance in relation to the 
history of civilization. 

32. What, then, were the principal words that the 
barbarians learnt from Rome in this period which may 
be called the pagan or pre-Christian period.?^ One of 
the earliest, no doubt, was wine (Lat. vinum), and a few 

I See especially Kluge , Paul's Gnnidriss , p. 327 ff.; Pogat- 
scher, J^autlehre der griech., lat. it. roman. leluiworte im alt- 
englischen (Strassb. 1888). I give the words in their modern 
English forms, wherever possible. 

Latin Words. 5j 

other words connected with the cultivation of the vine 
and the drinking of wine such as Lat. calicem, OE. calic 
(Germ, kelch) 'a cup'. It is worth noting, too, that the 
chief type of Roman merchants that the Germanic people 
dealt with, were the caupones 'wine-dealers, keepers of 
wine-shops or taverns'; for the word German kaufen, 
OE. ceapian 'to buy' is derived from it,^as is also cheap, 
the old meaning of which was 'bargain, price'. (Cf. 
Cheapside). Another word of commercial significance is 
monger (fishmonger, ironmonger, costermonger), OE. 
mangere from an extinct verb mangian, derived from Lat. 
mango 'retailer'. Lat. moneta, pondo, and uncia were 
also adopted as commercial terms: OE. mynet 'coin, 
coinage', now mint; OE. pund, now pound; OE. ynce, 
now inch; the sound-changes point to very early borrow- 
ing. Other words from the Latin connected with com- 
merce and travel are: mile, anchor, punt (OE. punt from 
Lat. ponto) ; a great many names for vessels or receptacles 
of various kinds; I take some from Pogatscher's list^ and 
add the modern forms if the word is still living: cist 
(chest), hinn (bin), byden, bytt, cylle, omber or amber 
(amber), disc (dish), scutel, ore, cytel (kettle), mortere 
(mortar), earc (ark), etc. This makes us suspect a com- i 
plete revolution in the art of cooking food, an impression 
which is strengthened by such Latin loan-words as cook 
(OE. coc from coquus), kitchen (OE. cycene from coquina) 
and mill (OE. mylen from molina), as well as names for 
a great many plants and fruits which had not previously 
been cultivated in the north of Europe, such as pear, 
OE. cirs 'cherry', persoc 'peach' (the modern forms are 
later adoptions from the French), plum (OE. plume, 
from prunus), pea (OE. pise from pisum), cole {caul, kale, 
Scotch kail, from Lat. caulis), OE. ncep, found in the 

I 1. c. 122. Cf. also Kluge, p. 331. 

32 II- The Beginnings. 

second syllable of mod. turnip, from napus, beet (root), 
mint, pepper, etc. As military words, though not wanting, 
were not taken over in such great numbers as one might 
expect, we have now gone through the principal cate- 
gories of early loans from the Latin language, from which 
conclusions as to the state of civilization may be drawn. 
In comparing them with later loan-words from the same 
source we are struck by their concrete character. It was 
not Roman philosophy or the higher mental culture that 
impressed our Germanic forefathers; they were not yet 
ripe for that influence, but in their barbaric simplicity 
they needed and adopted a great many purely practical 
and material things, especially such as might sweeten 
everyday life. It is hardly necessary to say that the 
words for such things were learnt in a purely oral manner, 
as shown in many cases by their forms; and this, too, 
is a distinctive feature of the oldest Latin loans as op- 
posed to later strata of loan-words. They were also short 
words, mostly of one or two syllables, so that it would 
seem that the Germanic tongues and minds could not 
yet manage such big words as form the bulk of later 
loans. These early words were easy to pronounce and to 
remember, being of the same general type as most of 
the indigenous words, and therefore they very soon 
came to be regarded as part and parcel of the native 
language, indispensable as the things themselves which 
they symbolized. 

Chapter III. 

Old English. 

33. We now come to the first of those important 
historical events which have materially influenced the 
Enghsh language, namely the settlement of Britain by 
Germanic tribes. The other events of paramount im- 
portance, which we shall have to deal with in succes- 
sion, are the Scand^inavian invasion, the Npxman con- 
quest, and the revival of learning. A future historian 
will certainly add the spreading of the English language 
in America, Australia, and South Africa. But none of 
these can compare in significance with the first con- 
quest of England by the EngHsh, an event which was, 
perhaps, fraught with greater consequences for the future 
of the world in general than anything else in history. 
The more is the pity that we know so very little either of 
the people who came over or of the state of things they 
found in the country they invaded. We do not know 
exactly when the invasion began; the date usually given ' 
is 449, but Bede, on whose authority this date rests, 
wrote about three hundred years later, and much may 
have been forgotten in so long a period. Many consider- 
ations seem to make it more advisable to give a rather 
earlier date;^ however, as we must imagine that the 

I R. Thurneysen, Wann sind die Germane n nach England 
gekommen? in Eng. Studien 22, 163. 

Jespersen: English. 2nd ed. 7 

34 ni. Old English. 

invaders did not come all at once, but that the settlement 
took up a comparatively long period during which new 
hordes were continually arriving, the question of date is 
of no great consequence, and we are probably on the safe 
side if we say that after a long series of Germanic invasions 
the country was practically in their power in the latter 
half of the fifth century. 

, 34. Who were the invaders, and where did they come 
from? This, too, has been a point of controversy. Accord- 
ing to Bede, the invaders belonged to the three tribes 
of Angles, Saxons, and Jutes; and linguistic history corrob- 
orates his statement in so far as we have really three 
dialects, or groups of dialects: the Anglian dialects in the 
■■ North with two subdivisions, Northumbrian and Mercian, 
the Saxon dialects in tlie grea'ter part of the South, the 
most important of which was the dialect of Wessex 
(West-Saxon), and the Kentish dialect, Kent having 
been, according to tradition, settled by the Jutes. 
But when Bede points out the district now called 
Angel (German Angeln) in South Jutland (Slesvig) 
^ as the home of the Anglians, and identifies the Jutes 
with the inhabitants of Jutland, his views have oi 
late years been much contested.^ It is not necessary 
here to enter on this debatable ground; suffice it to 
say that neither the language of the Anglians nor that 
of the Kentish people is Danish or ^hows any signs 
of closer relationship with Danish than West - Saxon, 
so that if the settlers came from Angel and other parts of 
Jutland, these districts cannot then have been inhabited 
by the same Danish population that has lived there as 

I See especially A. Erdmann, i/der die heiviat unci den namen 
der Angel?i. Upsala 1890. — H. Moller, Anzeiger fiir deutsches 
altertum XXII, lagff. — G. Schiitte, Var Angleme Tyskere, in 
S0nderjydske aarbeger 1900. — O. Bremer, in Paul's Grundriss 
I 2 1 1 5 ff. , where other references will be found. 

The invaders. le 

far back as ascertained history reaches. The continental 
language that shows the greatest similarity to English, 
is Frisian, and it is interesting to note that Frisian has 
some points in common with Kentish and some with 
Anglian, some even with the northernmost division of 
the Anglian dialect, points in which these OE. dialects 
differ from literary West-Saxon. Kentish resembles more 
particularly West Frisian, and Anglian East Frisian^ 
facts which justify us in looking upon the Frisians as the 
neighbours and relatives of the English before their emi- 
gration from the continent. We may therefore speak of 
Tan Anglo-Frisian language, forming in some respects a 
J connecting link between German Saxon (Low German) 
[ on the one hand and Scandmavian, especially Danish, on 
V^the other. 

35. What language or what languages did the sett- 
lers find on their arrival in Britain.? The original popu- 
lation was Celtic; but what about the Roman conquest.'* 
The Romans had been masters of the country for cen- 
turies; had they not succeeded in making the native 
population learn Latin as they had succeeded in Spain 
and Gaul? Some years ago Pogatscher^ took up the 
view that they had succeeded, and that the Angles and 
Saxons found a Brito-Roman dialect in full vigour. Po- 
gatscher endorsed Wright's view that 'if the Angles and 
Saxons had never come, we should have been now a 
people talking a Neo-Latin tongue , closely resem- 
bling French.' But this view was very strongly 
attacked by Loth^, and Pogatscher, in a subsequent 

1 W. Heuser, Altfriesisches lesebuch 1903 p. i — 5, and Indo- 
germanische forschungen, Anzeiger XIV 29. 

2 Zur lautlehre der . . . lehnworte ifn Altenglischen 

3 Les mots latins aans les langues brittoniques . Paris 



36 III. Old English. 

article^ had to withdraw his previous theory, if not 
completely, yet to a great extent, so that he no longer 
maintains that Latin ever was the national language 
of Britain, though he does not go the length of saying 
with Loth that the Latin language disappeared from 
Britain when the Roman troops were withdrawn. The 
possibility is left that while people in the country spoke 
Celtic, the inhabitants of the towns spoke Latin or that 
some of them did. However this may be, the fact remains 
that the English found on their arrival a population 
speaking a different language from their own. Did that, 
then, affect their own language, and in what manner and 
to what extent.? 

36. In his 'Student's History of England' p. 31 Gardiner 
says *So far as British words have entered into the Eng- 
lish language at all, they have been words such as gown 
or curd, which are likely to have been used by women, or 
words such as cart or pony^ which are likely to have been 
used by agricultural labourers, and the evidence of lan- 
guage may therefore be adduced in favour of the view 
that many women and many agricultural labourers were 
spared by the conquerors.' Here, then, we seem to have 
a Celtic influence from which an important historical in- 
ference can be drawn. Unfortunately, however, not a 
single word of those adduced can prove anything of the 
kind. For gown is not an old Celtic word, but was taken 
over from French in the 14th century (mediaeval Latin 
gunna); curd, too, dates only from the 14 th century, 
whereas if it had been introduced from Celtic in the old 
period we should certainly find it in older texts; 'it is not 
certain what relation (if any) the Celtic words hold to I 

I Angelsachsen u?id Rommieji. Engl. Studien XIX 329 — 352 
(1894). See also MacGillivray, The Influence of Christimtity on 
the Vocabulary of Old Eftglish p. XI. 

Celtic words. 


the English' (N. E. D.). Cart is an Old Norse word; it is 
found in Celtic languages, but is there 'palpably a foreign 
word' (N. E. D.) introduced from English; and pony'^, 
finally, is Lowland Scotch powney from Old French pou- 
lenet 'a little colt', a diminutive of poulain 'a colt'. Simi- , 
larly, most of the other words of alleged Celtic origin are , 
either Germanic or French words which the Celts have 
borrowed from English, or else they have not been used 
in England more than a century or two; in neither of 
these cases do they teach us anything with regard to the 
relations between the two nationalities fifteen hundred 
years ago.^ The net result of modern investigation seems • 
to be that not more than half a dozen words did pass over 
into English from the Celtic aborigines [bannock, brock, ■ 
crock, dun, dry 'magician', slough). How may we account 
for this very small number of loans? Sweet^ says the 
reason was that 'the Britons themselves were to a great 
extent Romanized', a theory which we seem bound to 
abandon now (see above). Are we to account for it, as 
Lindelof does* from the unscrupulous character of the 

1 Skeat, Notes on English Etymology 224. 

2 Curse, OE. cursian, is often referred to Ir. cursagaim, but 
'no word of similar form and sense is known in Celtic' (N. E. D.) 
Cradle, OE. cradol, seems to be a diminutive of an old Germanic 
word meaning 'basket' (O. H. G. chratto). See also hog in N. E. D. 
Windisch, in the article quoted below, p. 38, thinks that the 
Germanic tun in English took over the meaning of Celtic dunum 
(Latin 'arx') on account of the numerous old Celtic names of 
places in -dunum; but in OE. tu?i had more frequently the meaning 
of ' enclosure, yard' (cf. Dutch tuin) , ' enclosed land round a dwell- 
ing', 'a single dwelling house or farm' (cf. Old Norse tun\ still 
in Devonshire and Scotland); it was only gradually that the word 
acquired its modern meaning of village or town , long after the 
influenze of the Celts must have disappeared. — Sloga?t, pibroch, 
clan, etc , are modern loans from Celtic. 

3 New English Gra?nmar § 607. 

4 Grunddragen a/ Engelska sfirakets historiska Ijud- ochforrn- 
Idra (Helsingfors 1895 p. 47) — an excellent little book. 

38 HI. Old English. 

conquest, the English having killed all those Britons who 
did not run away into the mountainous districts? The 
supposition of wholesale slaughter is not, however, neces- 
sary, for a thorough consideration of the general con- 
ditions under which borrowings from one language by 
another take place will give us a clue to the mystery.^ 
And as the whole history of the English language may be 
described from one point of view as one chain of borrow- 
ings, it will be as well at the outset to give a little thought 
to this general question. 

37. The whole theory of Windisch about mixed lan- 
guages turns upon this formula: it is not the foreign lan- 
guage a nation learns which is made into a mixed language, 
but its own native language becomes mixed under the 
influence of the foreign language. When we try to learn 
and talk a foreign language we do not intermix it with 
words taken from our own language; our endeavour will 
always be to speak the other language as purely as pos- 
sible, generally we are painfully conscious of every native 
word that we use in the middle of phrases framed in the 
other tongue. But what we thus avoid in speaking a foreign 
language we very often do in our own. One of Windisch's 
illustrations is taken from Germany in the eighteenth 
century. It was then the height of fashion to imitate 
everything French, and Frederick the Great prided him- 
self on speaking and writing good French. In his French 
writings one finds not a single German word, but whenever 
he wrote German, French words and phrases in the middle 
of German sentences abounded, for French was considered 
more refined, more distingue. Similarly, in the last re- 

I See especially Windisch, Zur theorie der mischsprachen 
unci lehnworter. Berichte iiber die verhdl. d. sachs. gesellsch. 
d. wissensch. XLIX. 1897 p. 101 ff. — G. Hempl, Language- 
Rivalry' and Speech- Differentiation in the Case of Race -Mix- 
ture. Trans, of the Amer. Philol. Association XXIX. 1898 p. 3off. 

Mixed Languages. ig 

mains of Cornish, the extinct Celtic language of Cornwall, 
numerous English loan-words occur, but the English did 
not mix any Cornish words with their own language, and 
the inhabitants of Cornwall themselves, whose native 
language was Cornish, would naturally avoid Cornish 
words when talking English, because in the first place 
English was considered the superior tongue, the language 
of culture and civilization, and second, the English would 
not understand Cornish words. Similarly in the Brittany 
of to-day, people will interlard their Breton talk with 
French words, while their French is pure, without any 
Breton words. We now see why so few Celtic words were 
taken over into English.^ There was nothing to induce 
the ruling classes to learn the language of the inferior 
natives; it could never be fashionable for them to show 
an acquaintance with that despised tongue by using 
now and then a Celtic word. On the other hand the 
Celt would have to learn the language of his masters, , 
and learn it well; he could not think of addressing 
his superiors in his own unintelligible gibberish, and 
if the first generation did not learn good English, 
the second or third would, while the influence they 
themselves exercised on English would be infinitesimal. 
— There can be no doubt that this theory of 
Windisch's is in the main correct, though we shall, 
perhaps, later on see instances where it holds good 
only with some qualification. At any rate we need look 
for no other explanation of the fewness of Celtic words in 

38. About 600 A. D. England was christianized, and 
the conversion had far-reaching linguistic consequences. 
We have no literary remains of the pre-Christian period, j 

I And so few Gallic words into French. 

40 ni. Old English. 

\ but in the great epic of Beowulf we see a strange niixture 
of pagan and Christian elements. It took a long time 
thoroughly to assimilate the new doctrine, and, in fact, 
much of the old heathendom survives to this day in the 
shape of numerous superstitions. On the other hand we 
must not suppose that people were wholly unacquainted 
with Christianity before they were actually converted, 
and linguistic evidence points to their knowing, and hav- 
ing had names for, the most striking Christian pheno- 
mena centuries before they became Christians themselves. 
One of the earliest loan-words belonging to this sphere 
is church, OE. cirice , cyrice , ultimately from Greek 
kuriakSn '(house) of the Lord* or rather the plural 

, kuriakd. It has been well remarked that *it is by 
no means necessary that there should have been a 
single kirika in Germany itself; from 313 onwards. 
Christian churches with their sacred vessels and 
ornaments were well - known objects of pillage to 
the German invaders of the Empire: if the first 
with which these made acquaintance, wherever sit- 
uated, were called kuriakd, it would be quite sufficient 
to account for their familiarity with the word.'^ They 
knew this word so well that when they became Christians 
they did not adopt the word universally used in the Latin 
church and in the Romance languages {ecclesia, eglise, 
chiesa, etc.), and the English even extended the signi- 
fication of the word church from the building to the con- 
gregation, the whole body of Christians. Minster, OE. 
mynster from monasterium, belongs also to the pre- 
Christian period. Other words of very early adoption 
were devil from diaholus, Greek didbolos, and angel, OE. 

I See the full and able article church in the N. E. D. We 
need not suppose, as is often done, that the word passed through 
Gothic, where the word is not found in the literature that has 
come down to us. 

Christianity, 4 1 

engel^ from angelus, Greek dggelos. But the great bulk of 
specifically Christian terms did not enter the language till 
after the conversion. 

39. The number of new ideas and things introduced 
with Christianity was very considerable, and it is inter- 
esting to note how the English managed to express them 
in their language. ^ In the first place they adopted a great 
many foreign words together with the ideas. Such words 
are apostle OE. apostol, disciple OE. discipul, which has 
been more of an ecclesiastical word in English than in 
other languages, where it has the wider Latin sense of 
'pupir or 'scholar', while in English it is more or less 
limited to the twelve Disciples of Jesus or to similar 
applications. Further, the names of the whole scale of 
dignitaries of the church, from the Pope, OE. papa, 
downwards through archbishop OE. ercebiscop, bishop 
OE. biscop, to priest OE. preost; so also monk OE. munuc, 
nun OE. nunna with provost OE. prafost (praepositus) and 
profost (propositus), abbot OE. abbod (d from the Romance 
form) and the feminine OE. abbudisse. Here belong also 
such obsolete words as sacerd 'priest', canonic 'canon', 
decan 'dean', ancor or ancra 'hermit' (Latin anachoreta). 
To these names of persons must be added not a few 
names of things, such as shrine OE. serin (scrinium), cowl 
OE. cugele (cuculla), pall OE. pcell or pell (pallium); regol 
or reogol '(monastic) rule', capitul 'chapter', mcssse 'mass', 
and offrian, in Old English used only in the sense of 
'sacrificing, bringing an offering'; the modern usage in 

1 See below, § 86, on the^ relation between the OE. and the 
modern forms. 

2 See especially H. S. MacGillivray , The Influeitce of 
Christianity on the Vocabulary of Old English (Halle 1902). 
I arrange his material from other points of view and must 
often pass the limits of his book, of which only one half has 

42 III. Old English. 

*he offered his friend a seat and a cigar' is later and from 
the French. 

40. It is worth noting that most of these loans were 
short words that tallied perfectly well with the native 
words and were easily inflected and treated in every re- 
spect like these; the composition of the longest of them, 
ercebiscop, was felt quite naturally as a native one. Such 
long words as discipul or capitul, or as exorcista and acoli- 
tus, which are also found, never became popular words; 
and anachoreta only became popular when it had been 
shortened to the convenient ancor. 

41. The chief interest in this chapter of linguistic 
history does not, however, to my mind concern those 
words that were adopted, but those that were not. It 
is not astonishing that the English should have learned 
some Latin words connected with the new faith, but it 
is astonishing, especially in the light of what later gene- 
rations did, that they should have utilized the resources 
of their own language to so great an extent as was actu- 
ally the case. This was done in three ways : by forming 
new words from the foreign loans by means of 
native affixes, by modifying the sense of existing Eng- 
lish words, and finally by framing new words from 
native stems. 

At that period the English were not shy of affixing 
native endings to foreign words; thus we have a great 
many words in -had (mod. -hood): preosthad 'priesthood', 
clerichad, sacerdhad, hiscophad 'episcopate*, etc.; also such 
compounds as hiscopsedl 'episcopal see', hiscopscir 'dio- 
cese', and with the same ending profostscir 'provostship' 
and the interesting scriftscir 'parish, confessor's district' 
from scrift 'confession', a derivative of scrifan (shrive) 
which is the Latin scrihere with its signification curiously 
changed. Note also such words as cristendom 'Christen- 
dom, Christianity' (also cristnes), and cristnian 'christen' 

Native Words. 


or rather 'prepare a candidate for baptism'^ and biscopian 
'confirm' with the noun biscepung 'confirmation'. 

42. Existing native words were largely turned to ac- 
count to express Christian ideas, the sense only being 
more or less modified. Foremost among these must be 
mentioned the word God. Other wdrds belonging to the 
same class and surviving to this day are sin OE synn, 
tithe OE teoba, the old ordinal for 'tenth'; easier OE 
eastron was^,the name of an old pagan spring festival, 
called after Austro, a goddess of spring. ^ Most of the 
native words adapted to Christian usage have since been 
superseded by terms taken from Latin or French. Where 
we now say saint from the French, the old word was halig 
(mod. holy), preserved in All-hallows- day and Allhallow- 
e'en] the Latin sand was very rarely used. Scaru, from 
the verb scieran 'shear, cut' has been supplanted by 
tonsure, had by order, hadian by consecrate and ordain, 
gesomnung by congregation, ]>egnung by service, witega by 
prophet, 'prowere (irom J>rowian 'to suffer') by martyr, J>ro- 
werhad or prowung by martyrdom, niwcumen mann ('new- 
come man') by novice, hrycg-hrcegel (from hrycg 'back' 
and hrcegel 'dress') by dossal, and ealdor by prior. Com- 
pounds of the last-mentioned Old English word were also 
applied to things connected with the new religion, thus 
teobing- ealdor 'dean' (chief of ten monks). Ealdormann, 
the native term for a sort of viceroy or lord-lieutenant, 
was used to denote the Jewish High-Priests as well as the 
Pharisees. OE husl, mod. housel 'the Eucharist'^, was an 

1 ^Cristnian signifies primarily the 'prima signatio' of the 
catechumens as distinguished irom the baptism proper.' Mac 
Gillivray p. 2i. 

2 Connected with Sanscrit usra and Latin aurora and , there- 
fore, originally a dawn- goddess. 

3 Still used in the nineteenth century , e. g. by Tennyson , as 
an archaism. 

44 ni. Old English. 

old pagan word for sacrifice or offering; an older form 
is seen in Gothic hunsl. The OE word for 'altar', weofod, 
is an interesting heathen survival, for it goes back to a 
compound wigheod 'idol-table*, and it was probably only 
because phonetic development had obscured its connex- 
ion with wig 'idol' that it was allowed to remain in use 
as a Christian technical term. 

43. This second class is not always easily distinguished 
from the third, or those words that had not previously 
existed but were now framed out of existing native 
speech-material to express ideas foreign to the pagan 
world. Word-composition and other formative processes 
were resorted to, and in some instances the new terms 
were simply fitted together from translations of the com- 
ponent parts of the Greek or Latin word they were in- 
tended to render, as when Greek euaggelion was render- 
ed god-spell (good-spell, afterwards with shortening of 
the first vowel godspell, which was often taken to be the 
'speir or message of God), mod. gospel; thence godspellere 
where now the foreign word evangelist is used. Heathen, 
OE. hceben, according to the generally accepted theory, is 
derived from hcel> 'heath' in close imitation of Latin 
paganus from pagus 'a country district'. Of. also ^rynnes 
or prines ('three-ness') for trinity. 

44. But in most cases we have no such literal rendering 
of a foreign term, but excellent words devised exactly as 
if the framers of them had never heard of any foreign 
expression for the same conception — as, perhaps, in- 
deed, in some instances they had not. Some of these 
display not a little ingenuity. The scribes and Pharisees 
of the New Testament were called hoceras (from boc book) 
and sunder- halgan {irom sundor 'apart, asunder, separate'); 
in the north the latter were also called celarwas 'teachers 
of the Law' or celdo 'elders'. A patriarch was called 
heahfceder 'high-father' or eald- feeder 'old-father'; the 

New Terms. as 

three Magi were called tungol-witegan from tungol 'star', 
and witega 'wise man'. For 'chaplain' we have handpreost 
or hiredpreost ('family-priest') ; for 'acolyte' different word 
expressive of his several functions: husl^egn ('Eucharist- 
servant'), taporherend ('taper-bearer') and wcexberend 
('wax-bearer') ; instead of ercebiscop 'archbishop' we some- 
times find heahhiscop and ealdorbiscop. For 'hermit' 
ansetla and westensetla ('sole-settler', 'desert-settler') 
were used. 'Magic art' was called scincrceft ('phantom- 
art'); 'magician' scincrceftiga or scinlceca, scinnere, 'phan- 
tom' or 'superstition', scinlac. For the disciples of Christ 
we find, beside discipul mentioned above, no less than ten 
different English renderings (cniht, folgere, gingra, hiere- 
mon, Iceringman, leornere, leorning- cniht, leormngman, 
underpeodda, ^egn).^ To 'baptize' was expressed by 
dyppan 'dip' (cf. German taufen, Dan. debe) or more often 
by fulwian (from ful-wihan 'to consecrate completely'); 
'baptism' by fulwiht or, the last syllable being phoneti- 
cally obscured, fulluht, and John the Baptist was called 
Johannes se fulluhtere. 

45. The power and boldness of these numerous na- 
tive formations can, perhaps, best be appreciated if we 
go through the principal compounds of God: godbot 'ato- 
nement made to the church', godcund 'divine, religious, 
sacred*, godcundnes 'divinity, sacred office', godferht 
'pious', godgield 'idol', godgimm 'divine gem', godhad 
'divine nature', godmcegen 'divinity', godscyld 'impiety', 
godscyldig 'impious', godsibb 'sponsor', godsibbrceden 
'sponsorial obligations', godspell (cf., however, § 43), 
godspelbodung 'gospel-preaching', godspellere 'evangelist', 
godspellian 'preach the gospel', godspellisc 'evangelical', 
godspeltraht 'gospel-commentary', godsprcece 'oracle', god- 
sunu 'godson', god]>rymm 'divine majesty', godwrcec 'im- 

I MacGillivray p. 44.' 

^5 ni. Old English. 

pious', godwrcecnes 'impiety'. Such a list as this, with the 
modern translations, shows the gulf between the old 
system of nomenclature, where everything was native 
and, therefore, easily understood by even the most un- 
educated, and the modern system, where with few ex- 
ceptions classical roots serve to express even simple ideas; 
observe that although gospel has been retained, the easy 
secondary words derived from it have given way to learn- 
ed formations. Nor was it only religious terms that 
were devised in this way; for Christianity brought with 
it also some acquaintance with the higher intellectual 
achievements in other domains, and we find such scienti- 
fic terms as Icece-crceft ieech-craft' for medicine, tungol-ce 

, ('star-law') for astronomy, efnniht for equinox, sun-stede 
and sungihte for solstice, sunfolgend (sunfollower) for 
heliotrope, tid 'tide' and gemet 'measure' for tense and 

, mood in grammar, foresetnes for preposition etc., in short 
a number of scientific expressions of native origin, such 
as is equalled among the Germanic languages in Icelandic 

46. If now we ask, why did not the Anglo-Saxons adopt 
more of the ready-made Latin or Greek words, it is easy 
toseethattheconditions here are quite different from those 
mentioned above when we asked a similar question with 
regard to Celtic. There we had a real race-mixture, where 
people speaking two different languages were living in 
actual contact in the same country. Here we have no 
Latin-speaking nation or community in actual inter- 
course with the English; and though we must suppose 
that there was a certain mouth-to-mouth influence from 
missionaries which might familiarize part of the English 
nation with some of the specifically Christian words, 
these were certainly at first introduced in far greater num- 
ber through the medium of writing, exactly as is the case 
with Latin and Greek importations in recent times. Why, 

Why not Foreign Words? aj 

then, do we see such a difference between the practice 
of that remote period and our own time? One of the 
reasons seems obviously to be that people then did not 
know so much Latin as they learnt later, so that these 
learned words, if introduced, would not have been under- 
stood. We have it on King Alfred's authority that in the 
time immediately preceding his own reign 'there were 
very few on this side of the Humber who could under- 
stand their (Latin) rituals in English, or translate a 
letter from Latin into English, and I believe that there 
were not many beyond the Humber. There were so few 
of them that I cannot remember a single one south of 

the Thames when I came to the throne and there 

was also a great multitude of God's servants, but they 
had very Httle knowledge of the books, for they could 
not understand anything of them, because they were 
not written in their language.'^ And even in the previous 
period which Alfred regrets, when 'the sacred orders 
were zealous in teaching and learning', and when, as we 
know from Bede and other sources, ^ Latin and Greek 
studies were pursued successfully in England, we may 
be sure that the percentage of those who would have 
understood the learned words, had they been adopted 
into English, was not large. There was, therefore, good 
reason for devising as many popular words as possible. 
However, the manner in which our question was put was 
not, perhaps, quite fair, for we seemed to presuppose that 
it would be natural for a nation to adopt as many foreign 
terms as its linguistic digestion would admit, and that 
it would be matter for surprise if a language had fewer 

1 King Alfred's West -Saxon Version of Gregory's Pastoral 
Care. Preface (Sweet's translation). 

2 See T. N. Toller, Outlines of the History of the English 
Language. Cambridge 1900, p. 68fif. 

48 in. Old English. 

foreign elements than Modern English. But on the 
contrary, it is rather the natural thing for a language to 
utilize its own resources before drawing on other lan- 
guages. The Anglo-Saxon principle of adopting only 
such words as were easily assimilated with the native 
vocabulary, for the most part names of concrete things, 
and of turning to the greatest possible account native 
words and roots, especially for abstract notions, — that 
principle may be taken as a symptom of a healthful con- 
dition of a language and a nation; witness Greek, where 
we have the most flourishing and vigorous growth of 
abstract and other scientifically serviceable terms on a 
native basis that the world has ever seen, and where 
the highest development of intellectual and artistic 
activity went hand in hand with the most extensive 
creation of indigenous words and an extremely limited 
importation of words from abroad. It is not, then, 
the Old English system of utilizing the vernacular stock 
of words, but the modern system of neglecting the 
native and borrowing from a foreign vocabulary that 
'has to be accounted for as something out of the natural 
state of things. A particular case in point will illustrate 
this better than long explanations. 

47. To express the idea of a small book that is always 
ready at hand, the Greeks had devised the word egkhei- 
ridion from en 'in', kheir 'hand' and the suffix -idion 
denoting smallness; the Romans sirhilarly employed their 
adjective manualis 'pertaining to manus, the hand' with 
liber 'book' understood. What could be more natural 
then, than for the Anglo-Saxons to frame according to 
the genius of their own language the compound handboc? 
This naturally would be especially applied to the one 
kind of handy books that the clergy were in particular 
need of, the book containing the occasional and minor 
public offices of the Roman church. Similar compounds 



were used, and are used, as a matter of course, in the 
other cognate languages, — German handbuch, Danish 
handbog, etc. But in the Middle English period, handboc 
was disused, the French (Latin) manual taking its place, 
and in the sixteenth century the Greek word [enchiridion) 
too was introduced into the English language. And so 
accustomed had the nation grown to preferring strange 
and exotic words that when in the nineteenth century 
handbook made its re-appearance it was treated as 
an unwelcome intruder. The oldest example of the 
new use in the NED. is from 1814, when an anony- 
mous book was published with the title 'A Handbook 
for modelling wax flowers.' In 1833 Nicolas in the 
preface to a historical work wrote 'What the Germans 
would term and which, if our language admitted 
of the expression , would have been the fittest title 
for it, 'The Handbook of History' ', — but he dared 
not use that title himself. Three years later Murray 
the publisher ventured to call his guide-book 'A 
Hand - Book for Travellers on the Continent', but 
reviewers as late as 1843 apologized for copying this 
coined word. In 1838 Rogers speaks of the word 
as a tasteless innovation, and Trench in his 'Eng- 
lish Past and Present' (1854; 3rd ed. 1856 p. 71) 
says, 'we might have been satisfied with 'manual', and 
not put together that very ugly and very unnecessary 
word 'handbook', which is scarcely, I should suppose, 
ten or fifteen years old.' Of late years, the word seems 
to have found more favour, but I cannot help thinking 
that state of language a very unnatural one where such 
a very simple, intelligible, and expressive word has to 
fight its way instead of being at once admitted to the 
very best society. 

48. The Old English language, then, was rich in possi- 
bihties, and its speakers were fortunate enough to possess 

Jespersbn : English. 2nd ed. 4. 

50 HI. Old English. 

a language that might with very little exertion on their 
part be made to express everything that human speech 
can be called upon to express. There can be no doubt 
that if the language had been left to itself, it would 
easily have remedied the defects that it certainly had, 
for its resources were abundantly sufficient to provide 
natural and expressive terms even for such a new world 
of concrete things and abstract ideas as Christianity 
meant to the Anglo-Saxons. It is true that we often 
find Old English prose clumsy and unwieldy, but that 
is more the fault of the literature than of the language 
itself. A good prose style is everywhere a late acquire- 
ment, and the work of whole generations of good authors 
is needed to bring about the easy flow of written prose. 
Neither, perhaps, were the subjects treated of in the 
extant Old English prose literature those most suitable 
for the development of the highest literary qualities. 
But if we look at such a closely connected language 
as Old Norse, we find in that language a rapid 
progress to a narrative prose style which is even now 
justly admired in its numerous sagas; and I do not 
see so great a difference between the two languages 
as would justify a scepticism with regard to the 
perfectibility of Old English in the same direction. 
And, indeed , we have positive proof in a few passages 
that the language had no mean power as a literary 
medium; I am thinking of Alfred's report of the 
two Scandinavian travellers Ohthere and Wulfstan, 
who visited him — the Fridtjof Nansen and Sven 
Hedin of those days — , of a few passages in the Saxon 
Chronicle, and especially of some pages of the homilies 
of Wulfstan, where we find an impassioned prose of 
real merit. 

49. If Old English prose is undeveloped, we have a ; 
very rich and characteristic poetic literature, ranging 

Prose and Poetry. ei 

from powerful pictures of battles and of fights with 
mythical monsters to rehgious poems, idyllic descrip- 
tions of an ideal country and sad ones of moods of me- 
lancholy. It is not here the place to dwell upon the 
literary merit of these poems, as we are only concerned 
with the language. But to anyone who has taken the 
trouble — and it is a trouble — to familiarize himself 
with that poetry, there is a singular charm in the 
language it is clothed in, so strangely different from 
modern poetic style. The movement is slow and 
leisurely; the measure of the verse does not invite us 
to hurry on rapidly, but to linger deliberately on 
each line and pause before we go on to the next. 
Nor are the poet's thoughts too light-footed; he likes 
to tell us the same thing two or three times. Where 
a single he would suffice he prefers to give a couple 
of such descriptions as 'the brave prince, the bright 
hero, noble in war, eager and spirited' etc., descriptions 
which add no new trait to the mental picture, but 
which , nevertheless , impress us artistically and work 
upon our emotions , very much like repetitions and 
variations in music. These effects are chiefly produced 
by heaping synonym on synonym , and the wealth 
of synonymous terms found in Old English poetry is 
really astonishing, especially in certain domains, which 
had for centuries been the stock subjects of poetry. For 
'hero' or 'prince' we find in Beowulf alone at least thirty- 
seven words (cedeling. cescwiga. aglceca. headorinc. heag- 
gyfa. healdor. beorn. brego. brytta. byrnwiga. ceorl. cniht. 
cyning. dryhten. ealdor. eorl. ebelweard. jengel. frea. freca. 
fruma. hceleb. hlaford. hyse. lead. mecg. nid. oretta. rceswa. 
rinc. scota. secg. ^egn. jengel. peoden. wer. wiga). For 
'battle' or 'fight' we have in Beowulf at least twelve 
synonyms (beadu. gub. hea^o. hild. lindplega. nid. orleg. 
rcBs. sacu. geslyht. gewinn. wig). Beowulf has seventeen 


52 in. Old English. 

expressions for the 'sea' (brim. flod. garsecg. hcef. hea^u? 
holm, holmwylm. hronrad. lagu. mere, merestrcet. see. 
seglrad. stream, weed. wceg. yp), to which should be add- 
ed thirteen more from other poems (flodweg. flodwielm. 
fiot. flotweg. holmweg. hronmere. mereflod. merestream. 
sceflod. sceholm. scestream. sceweg. y]>mere). For 'ship' or 
'boat' we have in Beowulf eleven words (bat. brenting. 
ceol. jeer, flota. naca. scebat. scegenga. scewudu. scip. sund- 
wudu) and in other poems at least sixteen more words 
(brimhengest. brim^isa. brimwudu. cnearr. flodwudu. flot- 
scip. holmmcern. holmmcegen. merebat. merehengest. mere- 
])yssa. sceflota. scehengest. scemearh. ypbord. yphengest. 
y]>hof. yplid. yMida). 

50. How are we to account for this wealth of syno- 
nyms? We may subtract, if we like, such compound 
words as are only variations of the same comparison, as 
when a ship is called a sea-horse, and then different 
words for sea (see, mere, y]>) are combined with the 
words hengest 'stallion' and mearh 'mare'; but even if , 
this class is not counted, the number of synonyms is 
great enough to call for an explanation. A language ' 
has always many terms for those things that interest the 
speakers in their daily doings; thus Sweet says: 'if we 
open an Arabic dictionary at random, we may expect 
to find something about a camel: 'a young camel', 'an 
old camel', 'a strong camel', 'to feed a camel on the 1^ 
fifth day', 'to feel a camel's hump to ascertain its fat- 
ness', all these being not only simple words, but root- 
words'.^ And when we read that the Araucanians (in 
Chile) distinguished nicely in their languages between 
a great many shades of hunger, our compassion is ex- 
cited, as Gabelentz remarks.^ In the case of the Anglo- 

1 Sweet, The Practical Study of Language, 1899, p. 163. 

2 Gabelentz, Sprachwissenschaft 189, 463. - 

Synonyms. ^^ 

Saxons, however, the conclusion we are justified in 
drawing from their possessing such a great number of 
words connected with the sea is not , perhaps , that 
they were a seafaring nation, but rather, as these 
words are chiefly poetical and not used in prose , that 
the nation had been seafaring, but had given up that 
life while reminiscences of it were still Hngering in their 

51. In many cases we are now unable to see any 
difference in signification between two or more words, 
but in the majority of these instances we may assume 
that even if, perhaps, the Anglo-Saxons in historical 
times felt no difference, their ancestors did not use 
them indiscriminately. It is characteristic of primitive 
peoples that their languages are highly speciahzed , so 
that where we are contented with one generic word 
they have several specific terms. The aborigines of 
Tasmania had a name for each variety of gum-tree 
and wattle -tree, etc., but they had no equivalent for 
the expression 'a tree'. The Mohicans have words 
for cutting various objects, but none to convey cutting 
simply. The Zulus have such words as 'red cow', 
'white cow', 'brown cow', etc., but none for 'cow' 
generally. In Cherokee, instead of one word for 
'washing' we find different words, according to what 
is washed, 'I wash myself, — my head, — the head 
of somebody else, — my face, — the face of somebody 
else, — my hands or feet, — my clothes, — dishes, — a 
child, etc.^ 

52. Very Httle has been done hitherto to investigate 
the exact shades of meaning in Old English words, but 
I have Httle doubt that when we now render a number 

I Cf. Jespersen, Progress in Language, London 1894 
p. 250. 

54 in. Old English. 

of words indiscriminately by 'sword', they meant origi- 
nally distinct kinds of swords, and so in other cases as 
well. With regard to washing, we find something corre- 
sponding, though in a lesser degree, to the exuberance of 
Cherokee, for we have two words, wacsan (wascan) and 
iwean, and if we go through all the examples given in 
Bosworth and Toller's Dictionary, we find that the latter 
word is always applied to the washing of persons (hands, 
feet, etc.), never to inanimate objects, while wascan 
is used especially of the washing of clothes, but 
also of sheep , of 'the inwards' (of the victim, 
Leviticus I, 9 and 13^). Observe also that wascan was 
originally only used in the present tense (as Kluge infers 
from -sk-) , — a clear instance of that restriction in 
the use of words which is so common in the old stages 
of the language, but which so often appears unnatural 
to us. 

53. The old poetic language on the whole showed a 
great many divergences from everyday prose, in the 
choice of words, in the word forms, and also in the con- 
struction of the sentences. This should not surprise us, 
for we find the same thing everywhere, and the differ- 
ence between the dictions of poetry and of prose is 
perhaps greater in old or more primitive languages than 
in those most highly developed. In English, certainly, 
the distance between poetical and prose language was 
much greater in this first period than it has ever been 
since. The poetical language seems to have been to a 

I In a late text (R. Ben. 59, 7) we find the contrast agtier 
ge fata Jjwean, ge wcBterclacias wascan, which does not 
agree exactly with the distinction made above, — Curiously 
enough, in Old Norse, vaska is in the Sagas used only of 
washing the head with some kind of soap. In Danish, as 
well as in English, vaske, wash, is now the only word in 
actual use. 

Language of Poetry. 55 

certain extent identical all over England, regardless of 
dialect differences shown in prose writings. King Alfred's 
prose is always distinctly West Saxon, but when he 
breaks out occasionally into poetry, he uses such forms 
as the preterite heht, instead of het, the only form found 
in his prose. We have such more or less artificial poetic 
dialects, which agree with no one of the actually spoken 
dialects, in Homeric Greek and elsewhere, for example 
in the Old Saxon Heliand according to H. CoUitz.^ The, 
hypothesis of a poetical language of this kind, absorbing ' 
forms and words from the different parts of the country 
where poetry was composed at all, seems to me to offer 
a better explanation of the facts than the current theory, 
according to which the bulk of Old English poetry was 
written at first in Northumbrian dialect and later 
translated into West-Saxon with some of the old 
Anglian forms kept inadvertently — and translated 
to such an extent that no trace of the originals 
should have been preserved. The very few and short 
pieces extant in old Northumbrian dialect are easily 
accounted for, even if we accept the theory of a 
poetical koine or standard language prevailing in the 
time when Old English poetry flourished. But the whole 
question should be taken up by a more competent hand 
than mine. 

54. The external form of Old English poetry was in 
the main the same as that of Old Norse, Old Saxon, 
and Old High German poetry; besides definite rules of 
stress and quantity, which were more regular than might 
at first appear, but which were not so strict as those of 
classical poetry, the chief words of each line were tied 

I TAe Home of the Heliand; Publications of the Modern Lan- 
guage Association of America, Vol. XVI, p. I23fr. See also Bauer's 
Waldeckisches Worterbuch, 1901, p. 91* ff. 

^6 ni. Old English. 

together by alliteration, that is, they began with the 
same sound, or, in the case of sp, st, sc, with the same 
sound group. The effect is peculiar, and may be appre- 
ciated in such a passage as this: 

Him pa ellenrof andswarode, 
wlanc Wedera leod, word aefter spraec, 
heard under helme: 'We synt Higelaces 
beod-geneatas, Beowulf is min nama. 
Wille ic a-secgan suna Halfdenes, 
maerum peodne min aerende, 
aldre ]7inum gif he us geunnan wile, 
]73et we hine swa godne gretan mot on.' 
Wulfgar ma]7elode, Ipaet waes Wendla leod, 
waes his mod-sefa manegum gecy^ed, 
wig ond wisdom. 'Ic ]7ses wine Deniga, 
frean Scildinga, frinan wille, 
beaga bryttan, swa ]7u bena eart, 
]7eoden maerne ymb ]7inne sid.^ 

55. Very rarely, combined with alliteration we fird a 
sort of rime or assonance. In the prose of the last 
period of Old English the same artistic means were often 
resorted to to heighten the effect, and we find in Wulf- 
stan's homilies such passages as the following where all 
tricks of phonetic harmony are brought into play: 'in 
mordre and on mane, in susle and on sare, in wean 
and on wyrmslitum betweonan deadum and deoflum, 
in bryne and on biternesse, in bealewe and on bra- 
dum ligge, in yrm]?um and on earfe^um, on swyltcwale 
and sarum sorgum, in fyrenum bryne and on ful- 
nesse, in to^a gristbitum and in tintegrum' or again 
*)>aer is ece ece and J^aer is sorgung and sargung, 
and a singal heof ; ^aer is benda bite and dynta dyne, 
)?aer is wyrma slite and ealra wsedla gripe, ]>xr is wanung 

I Beowult 340 fif. 

Alliteration. cj 

and granung, J?aer is yrm^a gehwylc and ealra deofla 
gearing'. 1 

56. Nor has this love of alliterative word-combinations 
ever left the language; we find it very often in n^odern 
poetry, where however it is always subordinate to end 
rime, and we find it in such stock phrases as — : it can 
neither make nor war me, as ^usy as ^ees (Chaucer, 
E 2422), /?art and parcel, /aint and feeble, chucks and 
brakes (sometimes: play dick-duck-drake; Stevenson, 
Merry Men 277), what ain't missed ain't mourned (Pinero, 
Magistrate 5), as ^old as ^rass, free and /ranke (Caxton, 
Reynard 41), Barnes are Messings (Shakesp., All's I. 3. 28), 
as ^ool as a cucumber, as 5^ill as (a) stone (Chaucer, 
E 121, as any stoon E 171, he stode stone style, Malory 
145), over stile and stone (Chaucer B 1988), from top to 
^06 (from the top to toe, Shakesp. R 3 III. i. 155), 7;nght 
and main, fuss and /ume, manners makyth man, rare 
billed a rat, rack and ruin, nature and nurture (Shakesp. 
Tp. IV. I. 189; English Men of Science, their Nature 
and Nurture, the title of a book by Galton), etc. etc., 
even to Thackeray's 'faint fashionable fiddle-faddle and 
feeble court slipslop'. Alliteration sometimes modifies 
the meaning of a word, as when we apply chick to human f- 
offspring in 'no chick or child', or when we say 'a labour 
of love', without giving to labour the shade of meaning* 
which it generally has as different from work. The word' ' 
foe, too, which is generally used in poetry or archaic 
prose only, is often used in ordinary prose for the sake 
of aUitcration in connexion with /riend ('Was it an ir- 
ruption of a friend or a foe.?' Meredith, Egoist 439; 'The 
Danes of Ireland had changed from foes to friends', 
Green, Short Hist. 107). Indeed alHteration comes so 

I Wulfstan, Homilies, ed. by Napier, p. 187, 209. It is 
worthy of note that these poetical flights occur in descriptions 
of hell. 

58 III. Old English. 

natural to English people, that Tennyson says that 
'when I spout my lines first, they come out so allitera- 
tively that I have sometimes no end of trouble to get 
rid of the alliteration'.^ I take up the thread of my narra- 
tive after this short digression. 

I Life, by his Son, Tauchn. ed. II. 285. Cf. R. L. Stevenson, 
The Art of Writing 31, and what the Danish poet and metricist 
E. V. d. Recke says to the same effect, Principernefor den danske 
verskunst 1881, p. 112; see also the amusing note by De Quincey, 
Opium Eater p. 95 (Macmillan's Library of Eng. Classics): 'Some 
people are irritated, or even fancy themselves insulted, by overt 
acts of alliteration, as many people are by puns. On their 
account let me say, that, although there are here eight separate 
f's in less than half a sentence, this is to be held as pure accident. 
In fact, at one time there were nine fs in the original cast of 
the sentence, until I, in pity of the affronted people, substituted 
fe?na/e agent for female friend.' The reader need not be re- 
minded of the excessive use of alliteration in Euphuism and of 
Shakespeare's satire in Love's Labour's Lost and Midsufmner 
Night's Dream. 

Chapter IV. 

The Scandinavians. 

57. The Old English language, as we have seen, was 
essentially self-sufficing; its foreign elements were few 
and did not modify the character of the language as a 
whole. But we shall now consider three very important 
factors in the development of the language, three super- 
structures, as it were, that came to be erected on the 
Anglo-Saxon foundation, each of them modifying the 
character of the language, and each preparing the ground 
for its successor. A Scandinavian element, a French 
element, and a Latin element now enter largely into the 
texture of the English language, and as each element is 
characteristically different from the others, we shall 
treat them separately. First, then, the Scandinavian 

I The chief works on these loan-words, most of them treat- 
ing nearly exclusively phonetic questions, are: Erik Bjorkman, 
Scandinavian Loa7i-Wo7'ds in Middle English (Halle I 1900, 
II 1902), an excellent book; Erik Brate, No7dische Lehnworter 
im Orrmulum (Beitrage zur Gesch. d. deutschen Sprache X, 
Halle 1884); Arnold Wall, A Contribution towards the Study 
of the Scandinavia?i Ele??ieJit in the English Dialects (Anglia XX, 
Halle 1 8 98); G. T Flom, Scandinavian Influence on Southern 
Lowland Scotch (New York, 1900). The dialectal material of 
the two last -mentioned treatises is necessarily to a great extent 
of a doubtful character. See also Kluge in Paul's GrunariSs d. 
germ. Philol. 2nd ed. p. 931 ff. (Strassburg 1899J, Skeat, Principles 

5o IV. The Scandinavians. 

58. The EngHsh had resided for about four' centuries 
in the country called after them, and during that time 
they had had no enemies from abroad. The only wars 
they had been engaged in were internal struggles be- 
tween kingdoms belonging to, but not yet feeling them- 
selves as one and the same nation. ^The Danes were to 
them not deadly enemies but a brave nation from over 
the sea, that they felt to be of a kindred race with them- 
selves. The peaceful relations between the two nations 
may have been more intimate than is now generally 
supposed. Fresh light seems to be thrown on the sub- 
ject by the theory that an interesting, but hitherto 
mysterious Old English poem which is generally ascribed 
to the eighth century is a translation of a lost Scan- 
dinavian poem dealing with an incident in what was 
later to become the Volsunga Saga.^ This would establish 
a literary intercourse between England and Scandinavia 
previous to the Viking ages, and therefor^ would accord 
with the fact that the old Danish legends about 
'King Hrothgar and his beautiful hall Heorot'^were pre- 
served in England, even more faithfully than by the Danes 
themselves. Had the poet of Beowulf been able to 
foresee all that his countrymen werfi-deatiiied to suffer 
atJJifiJiands of the Danes, he would have chosen another 

subject for his great epic, and we should have missed 
the earliest noble outcome of the sympathy so often 
displayed by Englishmen for the fortunes of Denmark. 

of English Etymology p. 453 fif. (Oxford 1887), and some other 
works mentioned below. I have excluded doubtful material; but 
a few of the words I give as Scandinavian, have been considered 
as native by other writers. In most cases I have been convinced 
by the reasons given by Bjorkman. 

1 W. W. Lawrence, The First Riddle ofCyneiuulf; W. H. Scho- 
field, Signy's Lament. (Publications of the Modem Language 
Association of America, vol, XVII. Baltimore 1902.) 

Vikings. 6 1 

But as it is, in Beowulf no coming events cast their 
shadow before, and the English nation seems to have 
been taken entirely by surprise when about 790 the 
I long series of inroads began, in which 'Danes' and 'hea- 
V^thens' became synonyms for murderers and plunderers. 
At first the strangers came in small troops and disap- 
peared as soon as they had filled their boats with gold 
and other valuables; but from the middle of the ninth 
century, 'the character of the attack wholly changed. 
The petty squadrons which had till now harassed the 
coast of Britain made way for larger hosts than had as 
yet fallen on any country in the west; while raid and 
foray were replaced by the regular campaign of armies 
who marched to conquer, and whose aim was to settle 
on the land they won'.^ Battles were fought with various 
success, but on the whole the Scandinavians proved the 
stronger race and made good their footing in their new 
country. In the peace of Wedmore (878), King Alfred, 
the noblest and staunchest defender of his native soil, 
was fain to leave them about two-thirds of what we now 
call England; all Northumbria, all East Anglia and one 
half of Central England made out the district called the 

59. Still, the relations between the two races were ■ 
not altogether hostile. King Alfred not only effected 
the repulse of the Danes; he also gave us the first geo- 
graphical description of the countries that the fierce in- 
vaders came from, in the passage already referred to 
(§ 48). Under the year 959, one of the chroniclers says 
of the Northumbrian king that he was widely revered 
on account of his piety, but in one respect he was 
blamed : 'he loved foreign vices too much and gave heathen 


I J. R. Green, A Short History of the Engl. People, Illustr. 
ed. p. 87. 

62 IV. The Scandinavians. 

(i. e. Danish) customs a firm footing in this country, 
alluring mischievous foreigners to come to this land.' 
And in the only extant private letter in Old English^ 
the unknown correspondent tells his brother Edward 
that 'it is a shame for all of you to give up the English 
customs of your fathers and to prefer the customs of 
heathen men, who grudge you your very life; you show 
thereby that you despise your race and your forefathers 
with these bad habits, when you dress shamefully in 
Danish wise with bared neck and blinded eyes' (with 
, hair falling over the eyes.?). We see, then, that the 
English were ready to learn from, as well as to fight wnth 
the Danes. It is a small, but significant fact that in the 
glorious patriotic war-poem written shortly after the 
battle of Maldon (993) which it celebrates, we find for 
the first time one of the most important Scandinavian 
loan-words, to call] this shuais-iiaw: early the linguistic 
influenc e of the Danes began to be felt. 

60. A great number of Scandinavian families settled 
in England never to return, especially in Norfolk, Suffolk 
and Lincolnshire, but also in Yorkshire, Northumber- 
land, Cumberland, Westmoreland, etc. Numerous names 
of places, ending in-^y, -thorp (-torp), -beck, -dale, -thwaite, 
etc., bear witness to the preponderance of the invaders 
in great parts of England, as do also many names of per- 
sons found in English from about 1000 a. d.^ But these 
foreigners were not felt by the natives to be foreigners 
in the same manner as the English themselves had been 
looked upon as foreigners by the Celts. As Green has it, 
'when the wild burst of the storm was over, land, people, 
government reappeared unchanged. England still re- 

1 Edited by Kluge, Engl. Studien VIII, 62. 

2 Bjorkman, Nordische Personennamen in England (Halle 
19 10). 

Danish Settlements. 


mained Eng land; the conquerors sank quietly into the 
mass of those around them; and Woden yielded without 

' a struggle tQ^C hrist. The secret of this difference be- 
tween the two invasions was that the battle was no 
longer between men of different races. It was no longer 
a fight between Briton and German, between English- 
man and Welshman. The life of these northern folk was 
in the main the life of the earlier Englishmen. Their 
customs, their religion, their social order were the same; 
they were in fact kinsmen bringing back to an England 
that had forgotten its origins the barbaric England of 
its pirate forefathers. Nowhere over Europe was the 
fight so fierce, because nowhere else were the combatants 
men of one blood and one speech. But just for this reason 

^the fusion of the northmen with their foes was nowhere 
so peaceful and so complete.'^ — "^t should be remem- 
bered , too , that it was a Dane, Kin g Knut who 
achieved what every English ruler had failed to achieve, 
the union of the whole of England into one peaceful 
realm. ) 

61. King Knut was a Dane, and in the Saxon Chron- 
icle the invaders were always called Danes, but from 
other sources we know that there were Norwegians too 
among the settlers. Attempts have been made to de- 
cide by linguistic tests which of the two nations had the 
greater influence in England^, a question beset with 

1 J. R. Green, A Short History of the E. People, Illustr. ed. 
p. 84. 

2 Brate thought the loan-words exclusively Danish; Kluge, 
Wall, and Bjorkman consider some of them Danish, others 
Norwegian, though in details they arrive at different results. 
See Bjorkman, Zur dialektischen provenienz der nordischen 
lehnwdrter im Englischen, Sprakvetensk. sallskapets for- 
handlingar 1898 — 1901 , Upsala, and his Scand. Loan -Words 
p. 281 ff. 


IV. The Scandinavians. 

considerable difficulties and which need not detain us' 
here. Suffice it to say that some words, such as ME. boun, 
Mod. bound 'ready' (to go to), busk, boon, addle, point 
rather to a Norwegian origin, while others, such as -by 
in place-names, die (>), booth, drown, ME. sum 'as', 
agree better with Danish forms. In the great majority 
of cases, however, the Danish and Norwegian forms were 
at that time either completely or nearly identical, so 
that no decision as to the special homeland of the Eng- 
lish loans is warranted. In the present work I there- 
fore leave the question open, quoting Danish or ON 
(Old Norse, practically = Old Icelandic) forms according 
as it is most convenient in each case, meaning simply 

62. In order rightly to estimate the Scandinavian in- 
fluence it is very important to remember how great the 

Simij ari ty w?^s hp ^w^^^ ^^^'^ ^^iglj^l^ ?^^ ^'^^ ^(^rfiP To 

those who know only modern English and modern Danish, 
this resemblance is greatly obscured, first on account of 
the dissimilarities^ that are unavoidable when two 
nations live for nearly one thousand years with very 
little intercommunication, and when there is, accordingly, 
nothing to counterbalance the natural tendency towards ' 
differentiation, and secondly on account of a powerful 
foreign influence to which each nation has in the mean- i 
time been subjected, English from French, and Danish | 
from Low German. But even now we can see the essen- 
tial conformity between the two languages, which in I 
those times was so - much greater as each stood so much 
nearer to the common source. An enormous number 

I Bjorkman's final words are: 'These facts would seem to 
point to the conclusion that a considerable number of Danes 
were found everywhere in the Scandinavian settlements, while 
the existence in great numbers of Norwegians was confined to 
certain definite districts.' 

Similarities. 65 

of words were then identical in the two languages, so 
that we should now have been utterly unable to tell 
which language they had come from, if we had had no 
English literature before the invasion; nouns such as 
man, wife, father, mother, folk, house, thing, life, sor- 
row, winter, summer, verbs like will, can, meet, come, 
bring, hear, see, think, smile, ride, stand, still, sit, set, 
adjectives and adverbs Hke full, wise, well, better, best, 
mine and thine, over and under, etc. etc. The conse- 
quence was that an Englishman would have no great 
difficulty in understanding a viking, nay we have posi- 
tive evidence that Norse people looked upon the Eng- 
lish language as one with their own. In many cases, 
however, the words were already so dissimilar that it 
offers no difficulty to distinguish them, for instance, 
when they contained an original ai, which in OE. had 
become long a (OE. swan = ON. sveinn), or au, which 
in OE. had become ea (OE. leas = ON. lauss, louss), 
or sk, which in English became sh (OE. scyrte, now shirt 
= ON. skyrta). 

63. But there are, of course, many words to which 
no such reliable criteria apply, and the difficulty in de- 
ciding the origin of words is further complicated by the 
fact that the E nglish wo uld-oftf^n modify n word, when 
adoptingj t, acco rdi ng to so mejiiore or less^ vague feel- 
ing of the English sound that corresponded generally to 
this or that Scandinavian sound. Just as the name of 
the English king ^delred Eadgares sunu is mentioned 
in the Norse saga of Gunnlaugr Ormstunga, as A^alra^r 
Jatgeirsson, in the same manner shift is an Anglicized 
form of Norse skipta^; ON. brudlaup 'wedding' was modi- 
fied into hrydlop (cf. OE. bryd 'bride'; a consistent AngH- 
cizing would be hrydhleap) ; tidende is unchanged in Orrms 

I In ME. forms with sk are also found; Bjorkman p. 126. 

Jespersen: English, and ed. 5 

56 IV. The Scandinavians. 

ti^ennde, but was generally changed into tiding (s), cf. 
OE. tid and the common Eng. ending -ing; ON. ijdnusta 
'service' appears as ])eonest, Jjenest, and J)egnest; ON. 
words with the negative prefix u are made into English 
wn-, e. g. untime 'unseasonableness', unbain (ON. ubeinn) 
'not ready', unrad or unrcsd 'bad counsel'^; cf. also wcepna- 
getcBc below, and others. 

] 64. Sometimes the Scandinavians gave a fresh lease 
' of life to obsolescent or obsolete native words. The pre- 
pngifinn j^ for instance, is found only once or twice 
in OE. texts belonging to the pre-Scandinavian period, 
but after that time it begins to be exceedingly common 
in the North, from whence it spreads southward; it was 
used as in Danish with regard both to time and space 
and it is still so used in Scotch. Similarly ^<2^^ (OE. dcel) 
'appears to have been reinforced from Norse (dal), 
for it is in the North that the word is a living 
geographical name' (NED.) , and barn, Scotch bairn 
(OE. beam) would probably have disappeared in the 
North, as it did in the South, if it had not been 
strengthened by the Scandinavian word. The verb 
blend y too, seems to owe its vitality (as well as its 
vowel) to Old Norse , for blandan was very rare in Old 

65. We also see in England a phenomenon, which, I 
think, is paralleled nowhere else to such an extent, 
namely the existence side by side for a long time, 
sometimes for centuries, of two slightly differing forms 
for the same word, one the original English form and 
the other Scandinavian. In the following the first 
form is the native one, the form after the dash the 
imported one. 

I Though the Scand. form is also found in a few instances i 
oulist 'listless', oumautin 'swoon'. 

Parallel Forms. 


66. In some cases both forms survive in standard 
speech, though, as a rule, they have developed slightly 
different meanings: whole (formerly hool) — hale] both 
were united in the old phrase 'hail and hool' | no — nay\ 
the latter is now used only to add an amplifying remark 
('it is enough, nay too much'), but formerly it was used 
to answer a question, though it was not so strong a neg- 
ative as no ('Is it true? Nay/ 'Is it not true? No') | 
rear — raise \ from — fro, now used only in 'to and fro' | 
shirt — skirt \ shot — scot \ shriek — screak, screech \ edge 
— ^gg vb. (to egg on, 'to incite'). OE. leas survives 
only in the suffix -less (nameless, etc.), while the Scand. 
loose has entirely supplanted it as an independent word. 

67. In other cases, the Scandinavian form survives in 
dialects only, while the other belongs to the literary 
language: dew — dag 'dew, thin rain; vb. to drizzle' | 
true — trigg 'faithful, neat, tidy' | leap — loup \ neat — 
nowt 'cattle' | church — kirk^ \ churn — kirn^ \ chest — 
kist^ I mouth — mun \ yard — garth *a small piece of en- 
closed ground'. All these dialectal forms belong to Scot- 
land or the North of England. 

68. As a rule, however, one of the forms has in course 
of time been completely crowded out by the other. 
The surviving form is often the native form, as in the 
following instances : goat — gayte \ heathen — heythen, 
haithen \ loath — laith \ grey — gra, gro \ few — fa, fo 
ash(es) — ask \ fish — fisk \ naked — naken \ yarn — gam 
bench — bennk \ star — sterne \ worse — werre. Simil- 
arly the Scand. thethen, hethen, hwethen are generally 
supposed to have been discarded in favour of the native 
forms, OE. ^anon, heonan, hwanon, to which was added 
an adverbial s: thence, hence, whence-, but in reality these 
modern forms seem to be due to the Scandinavian ones, 

1 These >&-words are, however, subject to some doubt. 


58 IV. The Scandinavians. 

, whose vowels they keep; for the loss of th cf. since from 
^sithence (sithens, OE. si])j)an + s)^ 

I 69. This then leads us on to those instances in which 
the intruder succeeded in ousting the legitimate heir. 
Caxton in a well-known passage gives us a graphic des- 
cription of the struggle between the native ey and the 
Scandinavian egg: 

And certaynly our langage now used varyeth ferre 
from that whiche was used and spoken whan I was 
borne. For we englysshe men ben borne under the 
domynacyon of the mone, whiche is never stedfaste, 
but ever waverynge, wexynge one season, and waneth 
& dyscreaseth another season. And that comyn eng- 
lysshe that is spoken in one shyre varyeth from a 
nother. In so moche that in my dayes happened that 
certayn marchauntes were in a shippe in tamyse, for 
to have sayled over the see into zelande. And for 
lacke of wynde, thei taryed atte forlond, and wente 
to lande for to refreshe them. And one of theym 
named sheffelde,^ a mercer, cam in-to an hows and 
axed for mete; and specyally he axyd after eggys. 
And the goode wyf answerde, that she coude speke 
no frenshe. And the marchaunt was angry, for he 
also coude speke no frenshe, but wolde have hadde 
egges, and she understode hym not. And thenne at 
laste a nother sayd that he wolde have eyren. Then 
the good wyf sayd that she understod hym wel. Loo, 
what sholde a man in thyse dayes now wryte, egges 
or eyren. Certaynly it is harde to playse every man, 
by cause of dyversite & chaunge of langage.* 

Very soon after this was written, the old English 
forms ey, eyren finally went out of use. 

1 Probably a north. country man. 

2 Caxton's Efieydos, p. 2 — 3. (E. E. T, S. Extra Series 57. 

Native Words Discarded. 59 

70. Among other word-pairs similarly fated may be 
mentioned: OE. a, ME. 'ever' — ay (both were found 
together in the frequent phrase 'for ay and 00') | tho (cf. 
those) — they \ theigh, thah, theh and other forms — though \ 
swon — swain (boatswain, etc.) | tbirde — hirth \ eie — 
awe I Mnresdcei — Thursday \ in (on) I>e lijte — on lofte, 
now aloft I swuster — sister \ chetel — kettle; and finally 
not a few words with English y over against Scand. g: 
yete — get \ yeme 'care, heed' — goni(e), dialectal gaum 
'sense, wit, tact' | yelde — guild 'fraternity, association' |. 
yive or yeve — give \ yift — gift. In this last- mentioned 
word gift, not only is the initial sound due to Scandi- 
navian, but also the modern meaning, for the Old Eng- 
lish word meant 'the price paid by a suitor in consider- 
ation of receiving a woman to wife' and in the plural 
'marriage, wedding'. No subtler linguistic influence can 
be imagined than this, where a word has been modified 
both with regard to pronunciation and meaning, and 
curiously enough has by that process been brought 
nearer to the verb from which it was originally derived 

71. In some words the old native form has survived, 
but has adopted the signification attached in Scandi- 
navian to the corresponding word; thus dream in OE. 
meant 'joy', but in ME. the modern meaning of 'dream' 
was taken over from ON. draumr, Dan. drom; analogous 
cases are bread (OE. bread 'fragment'), bloom (OE. bloma 
'mass of metal'). In one word, this same process of sense- 
shifting has historical significance; the OE. eorl meant 
vaguely a 'nobleman' or more loosely 'a brave warrior' 
or 'man' generally; but under Knut it took over the mean- 
ing of the Norse jarl 'an under-king' or governor of 
one of the great divisions of the realm, thus paving the 
way for the present signification of earl as one of the 
grades in the (French) scale of rank. OE. freond meant 

yo IV. The Scandinavians. 

only 'friend', whereas ON. frcendi, Dan. frcende means 
'kinsman', but in Orrm and other ME. texts the word 
sometimes has the Scand. meaning^ and so it has to this 
day in Scotch and American dialects (see many instances 
in J. Wright's Dialect Dictionary, e. g. 'We are near 
friends, but we don't speak'); the Scotch proverb 'Friends 
agree best at a distance' corresponds to the Danish 
'Fraende er fraende vaerst'. OE. dwellan or dwelian meant 
only 'to lead astray, lead into error, thwart' or intr. 'to 
go astray'^; the intransitive meanings, 'to tarry, abide, 
remain in a place', which correspond with the Scandi- 
navian meanings, are not found till the beginning of the 
13th century. OE. ploh is found only with the meaning 
of 'a measure of land' (still in Scotch pleuch), but in ME. 
it came to mean the implement plough (OE. sulh) as in 
ON. pldgr. OE. holm meant 'ocean', but the modern 
word owes its signification of 'islet, flat ground by a 
river' to Scandinavian holm. 

72. These were cases of native words conforming to 
foreign speech habits; in other instances the Scandina- 
vians were able to place words at the disposal of the 
English which agreed so well with other native words 
as to be readily associated with them, nay which were 
felt to be fitter expressions for the ideas than the Old 
English words and therefore survived. Death (dea]?) and 
dead are OE. words, but the corresponding verbs were 
steorfan and sweltan; now it is obvious that Danish deya 
(now d&) was more easily associated with the noun and 
the adjective than the old verbs, and accordingly it was 

1 Saxon Chron. 11 35, which is given in the NED. as an in 
stance of this meaning, appears to me to be doubtful. 

2 Divelode, in /Elfric, Homilies i. 384, is wrongly trans- 
lated by Thorpe 'continued', so that Kluge is wrong as giving 
this passage as the earliest instance of the modern meaning; 
it means 'wandered, went astray'. 

Ready Associations. 7 1 

soon adopted (deyen, now die), while sweltan was dis- 
carded and the other verb acquired the more special signi- 
fication of starving. Scete, Mod. E. seat, was adopted be- 
cause it was at once associated with the verbs to sit and 
to set. The most important importation of this kind was! 
that of the pronominal forms they, them and their, which 
entered readily into the system of English pronouns be- 
ginning with the same sound (the, that, this) and were 
felt to be more distinct than the old native forms which 
they supplanted. Indeed these were liable to constant 
confusion with some forms of the singular number (he, 
him, her) after the vowels had become obscured, so that 
he and hie, him and heom, her (hire) and heora could no 
longer be kept easily apart. We thus find the obscured 
form, which was written a (or 'a), in use for *he' till the 
beginning of the i6th century (compare the dialectal 
use, for instance in Tennyson's 'But Parson a cooms an' 
a goas'), and in use for 'she' and for 'they' till the end 
of the 14 th century. Such a state of things would 
naturally cause a great number of ambiguities; but al- 
though the th-iorms must consequently be reckoned a 
great advantage to the language, it took a long time 
before the old forms were finally displaced, nay, the 
dative hem still survives in the form 'em ('take 'em'), 
which is now by people ignorant of the history of the 
language taken to be a shortened them; her 'their' is the 
only form for the possessive of the plural found in Chaucer 
(who says they in the nominative) and there are two or 
three instances in Shakespeare. One more Scandinavian 
pronoun is same, which was speedily associated with the 
native adverb same [swa same 'similarly'). Other words 
similarly connected with the native stock are want (adj. 
and vb.), which reminded the English of their own wan 
'wanting', wana 'want' and wanian 'wane, lessen', and ill, 
which must have appeared like a stunted form of evil, 

>j2 IV. The Scandinavians. 

especially to a Scotchman who had made his own devil 
into deil and even into ein. 

73. If now we try to find out by means of the loan- 
word test (see above, § 31) what were the spheres of hu 
man knowledge or activity in which the Scandinavians 
were able to teach the English, the first thing that strikes 
us is that the very earliest stratum of loan-words\ words 
which by the way were soon to disappear again from the 
language^, relate to war and more particularly to the 
navy: orrest 'battle', fylcian 'to collect, marshal', li^ 
•fleet', harda, cnear, scegt> different sorts of warships, ha 
'rowlock'. This agrees perfectly well with what the 
Saxon Chronicle relates about the English being inferior 
to the heathen in ship-building, until King Alfred under- 
took to construct a new kind of warships.^ 

74. Next, we find a great many Scandinavian law- 
terms; they have been examined by Professor Steenstrup 
in his well-known work on 'Danelag'.* He has there 
been able, in an astonishing number of cases, to show 
conclusively that the vikings modified the legal ideas of 
the Anglo-Saxons, and that numerous new law-terms 
sprang up at the time of the Scandinavian settlements 
which had previously been utterly unknown. Most of 
them were simply the Danish or Norse words, others 
were Anglicizings, as when ON. vapnatak was made into 
wcepnagetcBC (later wapentake) or when ON. heimsokn ap- 
pears as hamsocn 'house-breaking or the fine for that 
offence', or saklauss as sacleas 'innocent'. The most im- 
portant of these juridical imports is the word law itself, 

1 See Bjorkman, p. 5. 

2 They were naturally supplanted py French words , see below. 

3 Therefore, I cannot believe that ON. bat is a loan from 
OE lai (boat), although it is difficult to account for the vowel 
by any other theory. 

4 Copenhagen 1882 (— Normanneme IV). 

Legal Terms. ^^ 

known in England from the loth century in the form 
lagu, which must have been the exact Scandinavian form 
as it is the direct fore-runner of the ON. form log, ODan. 
logh.'^ By-law is now felt to be a compound of the pre- 
position by and law, but originally by was the Danish by 
'town, village' (found in Derby, Whitby, etc.), and the 
Danish genitive-ending is preserved in the other English 
form byrlaw. Other words belonging to this class are 
nicfing 'criminal, wretch', thriding 'third part', preserved 
in the mutilated form riding^, carlman 'man' as opposed 
to woman, bonda or bunda 'peasant', lysing 'freedman', 
^rcell, Mod. thrall,, mal 'suit, agreement', wi]>ermal 
'counter-plea, defence', seht 'agreement', stefnan 'summon', 
crafian now crave, landcop or anglicized landceap and 
lahcop or lahceap (for the signification see Steenstrup 
p. 192 ff.); ran 'robbery'; infangen]>eof later infangthief 
'jurisdiction over a thief apprehended within the manor'. 
It will be seen that with the exception of law, bylaw, 
thrall and crave — the least juridical of them all — these 
Danish law-terms have disappeared from the language 
as a simple consequence of the Norman conquerors taking 
into their own hands the courts of justice and legal 
affairs generally. Steenstrup's research, which is largely 
based on linguistic facts, may be thus summarized. The 
Scandinavian settlers re-organized the administration of 
the realm and based it on a uniform and equable division 
of the country; taxes were imposed and collected after 
the Scandinavian pattern; instead of the lenient criminal 

1 The OE. word was ce or csw , which meant 'marriage' as 
well and was restricted to that sense in late OE,, until it was 
displaced by the French word. 

2 North -thriding being heard as North-riding; in the case 
of the two other ridings of Yorkshire, East -thriding and West- 
thriding, the th-so\m.d was assimilated to the preceding /, the 
result in all three cases being the same misdivision of the word 
(' metanalysis '). 

y^. IV. The Scandinavians. 

law of former times, a virile and powerful law was intro- 
duced which was better capable of intimidating fierce and 
violent natures. More stress was laid on personal honour, 
as when a sharp line was drawn between stealthy or 
clandestine crimes and open crimes attributable to ob- 
stinacy or vindictiveness. Commerce, too, was regulated 
so as to secure trade. ^ 

75. Apart from these legal words it would be very 
difficult to point out any single group of words belonging 
to the same sphere from which a superiority of any des- 
cription might be concluded. Window is borrowed from 
vindauga ('wind-eye'); but we dare not infer that the 
northern settlers taught the English anything in archi- 
tecture, for the word stands quite alone; besides OE. 
had another word for 'window', which is also based on 
the eye-shape of the windows in the old wooden houses: 
eag^yrel 'eye-hole' (cf. nospyrel nostril.) ^ Nor does the 
borrowing of steak, ME. steyke from ON. steik prove any 
superior cooking on the part of the vikings. But it is 
possible that the Scandinavian knives (ME. knif from 
Scand. knif) were better than or at any rate different from 
those of other nations, for the word was introduced into 
French (canif) as well as into English. 

76. If, then, we go through the lists of loan-words, 
looking out for words from which conclusions as to the 
state of culture of the two nations might be drawn, we 
shall be doomed to disappointment, for they all seem to 
denote objects and actions of the most commonplace 
de xription and certainly do not represent any new set 

1 Steenstrup, Danelag p. 391 ff 

2 Most European languages use the \^2X. fenestra {G. fenster, 
Dutch venster, Welsh ^enester), which was also imported from 
French into English as fenester , in use from 1290 to 1548. 
Slavonic languages have okno, derived from oko 'eye'. On the 
eye -shape of old windows see R. Meringer, Indogerm. For- 
schungen XVI 1904, p. 125. 

Commonplace Words, 75 

of ideas hitherto unknown to the people adopting them. 
We find such everday nouns as husband, fellow, sky, 
skull, skin, wing, haven, root, skill, anger, gate^, etc. Among 
the adjectives adopted from Scand. we find meek, low, 
scant, loose, odd^, wrong, ill, ugly, rotten. The impression 
produced perhaps by this list that only unpleasant ad- 
jectives came into English from Scandinavia, is easily 
shown to be wrong, for happy and seemly too are derived 
from Danish roots, not to speak of stor, which was com- 
mon in Middle English for 'great', and dialectal adjec- 
tives like glegg 'clear-sighted, clever', heppen 'neat, tidy', 
gain 'direct, handy', (Sc. and North E. the gainest way, 
ON. hinn gegnsta veg, Dan. den genneste vej). The 
only thing common to the adjectives, then, is seen to be 
their extreme commonplaceness, and the same impression 
is confirmed by the verbs, as for instance, thrive, die, 
cast, hit, take, call, want, scare, scrape, scream, scrub, scowl, 
skulk, bask, drown, ransack, gape, guess (doubtful), etc. To 
these must be added numerous words preserved only in 
dialects (north country and Scotch) such as lathe 'barn' 
Dan. lade, hoast 'cough' Dan. hoste, flit 'move' Dan. flytte, 
gar 'make, do' Dan. gore, lait 'search for' Dan. lede, red 
up 'to tidy' Dan. rydde op, keek in 'peep in', ket 'carrion, 
horseflesh, tainted flesh, rubbish', originally 'flesh, meat' 
as Dan. kQd, etc., all of them words belonging to the 
same familiar sphere, and having nothing about them 
that might be called technical 01 indicative of a higher 
culture. The same is true of that large class of words 
which have been mentioned above (§ 65 — 72), where the 
Scandinavians did not properly bring the word itself, 

1 Gate 'way, road, street', frequent in some northern towns 
An the names of streets , frequent also in ME- adverbial phrases 

algate, anothergate{s) (corrupted into anotherguess), etc. In 
the sense 'manner of going' it is now spelt gait. 

2 Cf. North-Jutland dialect (Vendsyssel) oj 'odd (number)'. 

7 5 IV. The Scandinavians. 

but modified either the form or the .ignification of a 
native word; among them we have seen such everyday 
words as get, give, sister, loose, birth, awe, bread, dream, 
etc.^ It is precisely the most indispensable elements of^ 
the language that have undergone the strongest Scandi- 
navian influence, and this is raised into certainty whenl 
we discover that a certain number of those grammatical] 
words, the small coin of language, which Chinese gram- 
marians term 'empty words', and which are nowhere else 
transferred from one language to another, have been 
taken over from Danish into English: pronouns like 
they, them, their, the same and probably both', a modal 
verb like Scotch maun, mun (ON. munu, Dan. mon, 
monne); comparatives like minne 'lesser', min 'less', helder 
•rather'; pronominal adverbs like hethen, thethen, whethen 
'hence, thence, whence', samen 'together'; conjunctions 
like though, oc 'and', sum, which for a long time seemed 
likely to displace the native swa (so) after a comparison, 
until it was itself displaced by eallswa > as; prepositions 
L like fro and till (see above § 64). ^ 

77. It is obvious that all these non-technical words 
can show us nothing about mental or industrial superi- 
ority; they do not bear witness to the currents of civili- 
zation; what was denoted by them cannot have been 
new to the English; we have here no new ideas, only 
new names. Does that mean, then, that the loan-word 
test which we are able to apply elsewhere, fails in this 
one case, and that linguistic facts can tell us nothing 

1 It is noticeable, too, that the native word heaven has been 
more and more restricted to the figurative and religious accep- 
tation, while the Danish sky is used exclusively of the visible 
firmament; sky originally meant cloud. 

2 Another preposition, umbe, was probably to a large extent 
due to Scandinavian, the native form being ymde, embe\ but 
perhaps in some texts u in umbe may represent the vowel [y]. 

Intimate Fusion. *j<n 

about the reciprocal relations of the two races? No; 
on the contrary, the suggestiveness of these loans leaves 
nothing to be desired, they are historically significant 
enough. If the English loan-words in this period extend 
to spheres where other languages do not borrow, if the 
Scandinavian and the English languages were woven 
more intimately together, the reason must be a more in- 
timate fusion of the two nations than is seen anywhere 
else. They fought like brothers and afterwards settled 
down peaceably, like brothers, side by side. The num- 
bers of the Danish and Norwegian settlers must have 
been considerable, else they would have disappeared 
without leaving such traces in the language. 

78. It might at the first blush seem reasonable to 
think that what was going on among Scandinavian sett- 
lers in England was parallel to what we see going on 
now in the United States. But there is really no great 
similarity between the two cases. The language of Scan- 
dinavian and other settlers in America is often a curious 
mixture, but it is very important to notice that it is 
i D anish or Norwegian, spr in kled wi th English words: 
'han har fencet sin farm og venter en god krop' he has 
fenced his farm and expects a good crop; 'lad os krosse 
streeten' let us cross the street, 'tag det trae' take that 
tray; 'hun suede ham i courten for 25 000 daler' etc. 
But this is toto ccbIo different from the English language 
of the middle ages. And if we do not take into account 
those districts where Scandinavians constitute the im- 
mense majority of the population and keep up their old 
speech as pure as circumstances will permit, the children 
or at any rate the children's children of the immigrants 
speak English, and very pure English too without any 
Danish admixture. The English language of America 
has no loan-words worth mentioning from the languages 
of the thousands and thousands of Germans, Scandina- 


IV. The Scandinavians. 

vians, French, Poles and others that have settled there. 
Nor are the reasons far to seek.^ The immigrants come 
in small groups and find their predecessors half, or more 
than half, Americanized; those belonging to the same 
country cannot, accordingly, maintain their nationality i 
collectively; they come in order to gain a livelihood, | 
generally in subordinate positions where it is important 
to each of them separately to be as little different as 
possible from his new surroundings, in garb, in manners, 
and in language. The faults each individual commits 
in talking English, therefore, can have no con&jquences 
of lasting importance, and at any rate his children are 
in most respects situated like the children of the natives J 
and learn the same language in essentially the same 
manner. In old times, of course, many a Dane in Eng- 
land would speak his mother-tongue with a large admix- 
ture of English, but that has no significance in linguistic 
history, for in course of time the descendants of the im- 
migrants would no longer learn Scandinavian as their 
mother-tongue, but English. But that which is import- 
ant, is the fact of the English themselves intermingling 
their own native speech with Scandinavian elements. 
Now the manner in which this is done shows us that 
the culture or civilization of the Scandinavian settlers 
cannot have been of a higher order than that of the 
English, for then we should have seen in the loan-words 
special groups of technical terms indicative of this superi- 
ority. Neither can their state of culture have been much 
inferior to that of the English, for in that case they would 
have adopted the language of the natives without ap- 

I See G. Hempl's valuable paper on Language -Rivalry 
and Speech - Differentiation in the case of Race Mixture. (Trans- 
act, of the Amer. Philol. Association, XXIX, 1898, p. 35). 
Hempl's very short mention of the Scandinavians in England, 
is , perhaps , the least satisfactory portion of his paper ; none of 
his classes apply to our case. 

Speech Mixture. nq 

preciably influencing it. This is what happened with 
the Goths in Spain, with the Franks in France and with 
the Danes in Normandy, in all of which cases the Ger- 
manic tongues were absorbed into the Romance lan- 
guages.^ It is true that the Scandinavians were, for a 
short time at least, the rulers of England, and we have 
found in the juridical loan-words linguistic corroboration 
of this fact; but the great majority of the settlers did 
not belong to the ruling class. Their social standing must 
have been, on the whole, slightly superior to the average 
of the English, but the difference cannot have been 
great, for the bulk of Scandinavian words are of a purely 
democratic character. This is clearly brought out by a 
comparison with the French words introduced in the 
following centuries, for here language confirms what 
history tells us, that the French represent the rich, the 
ruling, the refined, the aristocratic element in the Eng- 
lish nation. How different is the impression made by 
the Scandinavian loan-words. They are homely ex- 
pressions for things and actions of everyday importance; 
their character is utterly democratic. The difference is 
also shown by so many of the French words having 
never penetrated into the speech of the people, so that 

I It is instructive to contrast the old speech - mixture in Eng- 
land with what has been going on for the last two centuries 
in the Shetland Islands. Here the old Norwegian dialect (' Norn ') 
has perished as a consequence of the natives considering it 
more genteel to speak English (Scotch). All common words 
of their speech now are English , but they have retained a certain 
number of Norn words, all of them technical, denoting different 
species of fish, fishing implements, small parts of the boat or 
of the house and its primitive furniture, those signs in clouds, 
i etc., from which the weather was forecast at sea, technicalities 
of sheep rearing, nicknames for things which appear to them 
ludicrous or ridiculous, etc. — all of them significant of the 
language of a subjugated and poor population. (J. Jakobsen, 
Det norr^me sprog pa Shetland, Copenhagen 1897.) 

3o IV. The Scandinavians. 

they have been known and used only by the 'upper 
ten', while the Scandinavian ones are used by high and 
low alike; their shortness too agrees with the mono- 
syllabic character of the native stock of words, conse- 
quently they are far less felt as foreign elements than^ 
many French words; in fact, in many statistical calcu- 
lations of the propoition of native to imported words in' 
English, Scandinavian words have been more or less in- 
advertently included in the native elements. Just as 
it is impossible to speak or write in English about higher 
intellectual or emotional subjects or about fashionable 
mundane matters without drawing largely upon the 
French (and Latin) elements, in the same manner Scan- 
dinavian words will crop up together with the Anglo- 
Saxon ones in any conversation on the thousand nothings 
of daily life or on the five or six things of paramount 
importance to high and low alike. An Englishman cannot 
thrive or be ill or die without Scandinavian words; they 
are to the language what bread and eggs are to the daily 
fare. To this element of his language an Englishman 
might apply what Wordsworth says of the daisy: 

Thou unassuming common -place 
Of Nature, with that homely face 
And yet with something of a grace 
Which Love makes for thee! — 

79. The form in which the words were borrowed oc- 
casions very few remarks. Those nouns which in Scand. 
had the nominative ending -r, did not keep it, the kernel 
only of the word (= accus.) being taken over. In one 
instance the Norse genitive-ending appears in English; \ 
the Norse phrase d ndttar ]>eli 'in the middle of the night' ' 
(pel means 'power, strength') was Anglicized into on ; 
nighter tale (Cursor Mundi), or bi nighter tale (Havelock, ? 
* Chaucer etc.). The -t in neuters of adjectives, that 

Grammar. 8 1 

distinctive Scandinavian trait, is found in scani^, want ' 
and (a) thwart. Most Norse verbs have the weak inflexion 
in English, a3 might be expected {e. g. die, which in Old 
Scand. was a strong verb), but there is one noteworthy 
exception, take, that kept its Scand. strong inflection, 
ON. taka tdk taken. There are a few interesting words 
with the Scand. passive voice in -sk (from the reflexive 
pronoun sik): bask^ and busk^, but in English they are 
treated like active forms. The shortness of the ^^-forms 
may have led to their being taken over as inseparable 
wholes, for ON. otlask and privask lost the reflexive end- 
ing in English addle 'acquire, earn' and thrive. 

80. As the Danes and the English could understand 
one another without much difficulty it was natural that 
many niceties of grammar should be sacrificed, the in- 
telligibility of either tongue coming to depend mainly 
on its mere vocabulary.* So when we find that the 
wearing away and leveUing of grammatical forms in the 
regions in which the Danes chiefly settled was a couple 
of centuries in advance of the same process in the more 
southern parts of the country, the conclusion does not 
seem unwairantable that this is due to the settlers who 
did not care to learn EngHsh correctly in every minute 
particular and who certainly needed no such accuracy 
in order to make themselves understood. 

80 a. With regard to syntax our want of adequate early 
texts in Scandinavia as well as in North England makes 

I Properly skammt, neuter of skammr 'short'; the derived 
verb skemta, Dan. skemte 'joke' is found in ME. skemten. 
' 2 ON. bdba-sk 'bathe oneself rather than baka-sk 'bake 

3 ON. bua-sk 'prepare oneself. 

4 Jespersen, Progress iii Language, p. 173. Compare the ex- 
planation of the similar simplification of Dutch in South Africa 
given by H.Meyer, Die Sprache der Buren. (Gottingen 1901, p. 16.) 

Jespersen: English. 2nd ed. 6 

g2 / IV, The Scandinavians. 

it impossible for us to state anything very definite; but 
the nature of those loans which we are able to verify, 
warrants the conclusion that the intimate fusion of the 
two languages must certainly have influenced syntactical 
relations, and when we find in later times numerous 
striking correspondences between English and Danish, 
it seems probable that some at least of them date from 
the viking settlements, i It is true, for instance, that rela- 
tive clauses without any pronoun are found in very rare 
cases in Old English; but they do not become common 
till the Middle English period, when they abound; the 
use of these clauses is subject to the same restrictions 
in both languages, so that in ninety out of a hundred 
instances where an Englishman leaves out the relative 
pronoun, a Dane would be able to do likewise, and vice 
versa. The rules for the omission or retention of the 
conjunction that are nearly identical. The use of will 
and shall in Middle English corresponds pietty nearly 
with Scandinavian; if in Old English an auxiliary was 
used to express futurity, it was generally sceal, just as in 
modern Dutch (sal) ; wile was rare. In Modern Enghsh 
the older rules have been greatly modified, but in many 
cases where English commentators on Shakespeare note 
divergences from modern usage, a Dane would have 
used the same verb as Shakespeare. Furness, in his 
note to the sentence 'Besides it should appear' (Merch. 
III. 2. 289 = 275 Globe ed.) writes: 'It is not easy to 
define this 'should' .... The Elizabethan use of should 
is to me always difficult to analyse. Compare Stephano's 
question about Caliban: 'Where the devil should he learn 
our language .f*' Now, a Dane would say 'det skulde, 
synes', and 'Hvor fanden skulde han laere vort sprog?' 
Abbott (Shakeip. Grammar § 319) says 'There is a diffi- 
culty in the expression 'perchance I will'; but, from its 
constant recurrence, it would seem to be a regular idiom';) 


Syntax. g^ 

a Dane, in the three quotations given, would say vil. And 
similarly in other instances. *He could have done it' 
agrees with 'han kunde have gjort det' as against 'er 
hatte es tun konnen' (and French *il aurait pu le faire'), 
and the Scotch idiom 'He wad na wrang'd the vera 
Deir (Burns), 'ye wad thought Sir Arthur had a pleasure 
in it' (Scott), where Caxton and the Elizabethans could 
also omit have, has an exact parallel in Danish 'vilde 
gjort', etc. Other points in syntax might perhaps be 
ascribed to Scandinavian influence, such as the universal 
position of the genitive case before its noun (where Old 
English like German placed it very often after it), the 
use of a preposition governing a dependent clause (he 
talked of how people had injured him, found as early 
as Orrmulum; here German must say davon, wie, and 
Dutch er van hoe), etc.; but in these delicate matters it 
is not safe to assert too much, as in fact many similarities 
may have been independently developed in both lan- 

Chapter V. 
The French. 

81. If with regard to the Scandinavian invasion histo- 
rical documents were so scarce that the linguistic evi- 
dence drawn from the number and character of the loan- 
words was a very important supplement to our histori- 
cal knowledge of the circumstances, the same cannot be 
said of the Norman Conquest. T lhe Normans^ m uch more 
^an the Danes, were felt as an alip n rarp; i-Viair occu- 
pation of the country attracted much more notice and 
lasted much longer; they became the ruling^class and 
as_^iidi_Jvere much more spoken of in contemporary 
literature and in historical records than the comparatively 
obscure Scandinavian element; and finally, they repre- 
sented a higher culture than the natives and had a 
literature of their own, in which numerous direct state- 
ments and indirect hints tell us about their doings and 
their relations with the native population. No wonder, 
therefore, that historians should have given much more 
attention to this fuller material and to all the interesting 
problems connected with the Noiman conquest than to 
the race-mixture attending the Scandinavian immigra- 
tions. This is true in respect not only of political andjj 
social history, but also of the language, in which the Nor- 
man-French element is so conspicuous, and so easily ac- 
cessible to the student that it has been discussed very!] 
often and from various points of view. And yet, there is 


The Rulers of England. 85 

still much work for future investigators to do. In accord- 
ance with the geneial plan of my work, I shall in this 
chapter deal chiefly with what has been of permanent 
importance to the future of the English language, and 
endeavour to characterize the influence exercized by 

(French as contrasted with that exercized by other lan- 
guages with which English has come into contact. 

82. The. Normans hfrc^mp mac;<-pr<; nf_Fng1anH^ and 

they remained masters for a sufficiently long time to 
leave a deep impress on the language. The conquerors 
were numerous and powerful, but the linguistic influence 
would have been far less if they had not continued for 
centuries in actual contact and constant intercourse with 
the French of France, of whom many were induced by 
later kings to settle in England. We need only go through 
a list of French loan-words in English to be firmly con- 
vinced of the fact that the immigr ants formed the upper 
classes of the English societyjifter the conquestTso^any 
>of the woxds,are_distinctly aristocratic^ It is true that 
they left the old words king and queen intact, but apart 
from these nearly all words relating to government and 
to the highest administration are French; see, for in- 
stance, crown, state, government and to govern, reign^ 
realm (0. Fr. realme, Mod. Fr. royaume), sovereign^ 
country, power; minister, chancellor, council (and counsel), 
authority, parliament, exchequer. People and nation, too, 
were political words; the corresponding OE. Jjeod is not 
found latei than the thirteenth century. Feudalism was 
imported from France, and with it were introduced a 
number of words, such as fief, feudal, vassal, liege, and 
[ the names of the various steps in the scale of rank: 
j prince, peer, duke with duchess, marquis, viscount, baron^ 
I It is, perhaps, surprising that lord and lady should have 
[ remained in esteem, and that earl should have been 
i retained, count being chiefly used in speaking of for- 

86 V. The French. 

eigners, but the earl's wife was designated by the French 
word countess, and court is French, as well as the ad- 
jectives relating to court life, such as courteous, noble, 
fine and refined. Honour and glory belong to the French, 
and so does heraldry, while nearly all English expressions 
relating to that difficult science are of French origin, 
some of them curiously distorted. 

83. The upper classes, as a matter of course, took into 
their hands the management of military matters; and 
although in some cases it was a long time before the old 
native terms were finally displaced {here and fird, for 
instance, were used till the fifteenth century when army 
began to be common), we have a host of French mili- 
tary words, many of them of very early introduction. 
Such are war (ME. werre. Old North Fr. werre, Central 
French guerre) and peace, battle, arms, armour, buckler, 
hauberk, mail (chain-mail; O Fr. maille 'mesh of a net'), 
lance, dart, cutlass, banner, ensign, assault, siege, etc. 
Further officer, colonel, chieftain {captain is later), lieu- 
tenant, sergeant, soldier, troops, dragoon, vessel, navy and 
admiral (orig. amiral in English as in French, ultimately 
an Arabic word). Some words which are now used very 
extensively outside the military sphere, were without 
any doubt at first purely military, such as challenge^ 
enemy, danger, escape (scape), espy (spy), aid, prison^ 
hardy, gallant, march, force, company, guard, etc. 

84. Another natural consequence of the power of the 
Norman upper classes is that most of the terms per- 
taining to the law are of French origin, such as justice, 
just, judge; jury, court (we have seen the word already 
in another sense), suit, sue, plaintiff and defendant, a plea, 
plead, to summon, cause, assize, session, attorney, fee, ac- 
cuse, crime, guile, felony, traitor, damage, dower, heritage, 
property, real estate, tenure, penalty, demesne, injury, priv 
ilege. Some of these are now hardly to be called tech- 


Military and Legal Words. 87 

nical juridical words, and there are others which belong 
still more to the ordinary vocabulary of every-day life, 
but which were undoubtedly at first introduced by lawyers 
at the time when procedure was conducted entirely in 
French^; for instance, case, marry, marriage, oust, prove^ 
false (pel haps also fault), heir, probably also male and 
female, while defend and prison are common to the juri- 
dical and the military worlds. Petty (Fr. petit) was, I 
suspect, introduced by the jurists in such combinations 
as petty jury, petty larceny, petty constable, petty sessions, 
petty averages, petty treason (still often spelt petit treason), 
etc., before it was used commonly. The French puis ne 
in its legal sense remains puisne in English (in law it 
means 'younger or inferior in rank', but originally 'later 
born*), while in ordinary language it has adopted the spell- 
ing puny, as if the -y had been the usual adjective ending. 
85. Besides, there are a good many words that have 
never become common property, but have been known 
to jurists only, such as mainour (to be taken with the 
mainour, to be caught in the very act of steahng, from 
Fr. manoeuvre), jeofail ('an oversight', the acknowledge- 
ment of an error in pleading, from je faille), cestui que 
trust, cestui (a) que vie and other phrases equally shrouded 
in mystery to the man in the street. Larceny has been 
almost exclusively the property of lawyers, so that it has 
not ousted theft from general use; such words as thief 
and steal were of course too popular to be displanted 
by French juridical terms, though burglar is probably of 
French origin. It is also worth observing how many of 
the phrases in which the adjective is invariably placed 

I From 1362 English was established as the official lan- 
guage spoken in the courts of justice , yet the curious mongrel 
language known as 'Law French' continued in use there for 
centuries; Cromwell tried to break its power, but it was not 
finally abolished till an act of Parliament of i73i- 

38 V. The French. 

after its noun, are law terms, taken over bodily from 
the French, e. g. heir male, issue male, fee simple, proof \ 
demonstrative, malice prepense (or, Englished, malice 
aforethought)'^, letters patent (formerly also with the ad- 
jective inflected, letters patents, Shakesp. R 2 II i. 202), 
attorney general (and other combinations of general, all of 
which are official, though some of them are not juridical). 

86. As ecclesiastical matters were also chiefly under 
the control of the higher classes, we find a great many 
French words connected with the ch\irch, such as religion, 
service, trinity, saviour, virgin, angel (O Fr. angele, now 
Fr. ange; the OE. word engel was taken direct from Latin, 
see § 38), saint, relic, abbey, cloister, friar (ME. frere as 
in French), clergy, parish, baptism, sacrifice, orison, homily, 
altar, miracle, preach, pray, prayer, sermon, psalter (ME. 
sauter), feast ('religious anniversary'). Words like rule, 
lesson, save, tempt, blame, order, nature, which now belong ' 
to the common language and have very extensive ranges 
of signification, were probably at first purely ecclesiasti- 
cal words. As the clergy were, moreover, teachers of 
morality as well as of religion they introduced the 
whole gamut of words pertaining to moral ideas from 
virtue to vice: duty, conscience, grace, charity, cruel, 
chaste , covet , desire , lechery, fool (one of the oldest 
meanings is 'sensual'), jealous, pity, discipline, mercy, 
and others. 

87. To these words, taken from different domains, may 
be added other words of more general meaning, which 
are highly significant as to the relations between the 
Normans and the English, such as sir and madam, master 
and mistress with their contrast servant (and the verb to 
serve), further command and obey, order, rent, rich and 

I Cf. also lords spiritual and lords temporal', the body 

Masters and Servants. 8q 

poor with the nouns riches and poverty; money, interest^ 
cash, rent, etc. 

88. It is a remark that was first made by John WaHis^ 
and that has been very often repeated, especially since 
Sir Walter Scott made it popular in 'Ivanhoe', that 
while the names of seveial animals in their lifetime are 
English (oXf cow, calf, sheep, swine, boar, deer) they 
appear on the table with French names (heef, veal, mutton, 
pork, bacon, brawn, venison). This is generally explained | 
from the masters leaving the care of the living animals 
to the lower classes, while they did not leave much of 1 
the meat to be eaten by them. But it may with just as i 
much right be contended that the use of the French 
words here is due to the superiority of the French cuisine, 
which is shown by a great many other words as well, 
such as sauce, boil, fry, roast, toast, pasty, pastry, soup, 
sausage, jelly, dainty; while the humbler breakfast is Eng- 
lish, the more sumptuous meals, dinner and supper, as 
well as feasts generally, are French. 

89. We see on the whole that the masters knew how 
to enjoy life and secure the best things to themselves; / 
note also such words as joy and pleasure, delight, ease' 
and comfort] flowers and fruits may be mentioned in the 
same category. And if we go through the different ob- 
jects or pastimes that make life enjoyable to people 
having plenty of leisure (this word, too, is French) we 
shall find an exceedingly large number of French words. 
The chase^ of course was one of the favourite pastimes, 
and though the native hunt was never displaced, yet 
we find many French terms relating to the chase, such 
as brace and couple, leash, falcon, quarry, warren, scent, 
track. The general term sport, too, is of course a French 

1 Grammatica linguae Anglicanae 1653. 

2 This is the Central French form of the word that was 
taken over in a North French dialectal form as catch (Latin captiare). 

go V. The French. 

word; it is a shortened form of desport (disport). Cards 
and dice are French words, and so are a great many 
words relating to different games (partner, suit, trump), 
some of the most interesting being the numerals used 
by card and dice players: ace, deuce, tray, cater, cinque, 
size; cf. Chaucer's 'Sevene is my chaunce, and thyn is 
cynk and treye' (C 653). 

90. The French led the fashion in the middle ages, 
just as they do to some extent even now, so we expect 
to find a great many French words relating to dress; 
in fact, in going through Chaucer's Prologue to the 
Canterbury Tales, where in introducing his gallery of 
figures he seldom omits to mention their dress, one will 
-see that in nearly all cases where etymologists have been 
.able to trace the special names of particular garments 
to their sources these are French. And of course, such 
general terms as apparel, dress, costume, and garment are 
derived from the same language. 

91. The French were the teachers of the English in 
most things relating to art; not only such words as art, 
beauty, colour, image, design, figure, ornament, to paint, but 
also the greater number of the more special words of 
technical significance are French; from architecture may 
be mentioned, by way of specimens: arch, tower, pillar, 
vault, porch, column, aisle, choir, reredos, transept, chapel, 
cloister (the last of which belong here as well as to our 
§ 86), not to mention palace, castle, manor, mansion, etc. 
If we go through the names of the various kinds of 
artisans, etc., we cannot fail to be struck with the dif- 
ference between the more homely or more elementary 
occupations which have stuck to their old native names 
(such as baker, miller, smith, weaver, saddler, shoemaker, 
wheelwright, -fisherman, shepherd, and others), on the one 
hand, and on the other those which brought their practi- 
tioners into more immediate contact with the upper 

Dress, Art, Phrases. gi 

classes, or in which fashion perhaps played a greater part; 
these latter have French names, for instance, tailor^ 
butcher, mason, painter, carpenter and joiner (note also 
such words as furniture, chair, table etc.). 

92. I am afraid I have tired the reader a little with 
all these long lists of words. My purpose was to give 
abundant linguistic evidence for the fact that the French 
were the rich, the powerful, and the refined classes. It 
was quite natural that the lower classes should soon begin 
to imitate such of the expressions of the rich as they 
could catch the meaning of. They would adopt inter- 
jections and exclamations like alas, certes, sure, adieu; 
and perhaps verray (later very) was at first introduced 
as an exclamation. Whole phrases were adopted: in 
the Ancrene Riwle (about 1225) we find (p. 268) Deu- 
leset (Dieu le sait) in two manuscripts while a third has 
j Crist hit wat; and three hundred years later, we find 'As 
good is a becke (= a wink), as is a dewe vow garde* (Bale, 
Three Lawes I. 1470). As John of Salisbury (Johannes 
Sarisberiensis) says expressly in the twelfth century^, 
it was the fashion to interlard one's speech with 
f French words; they were thought modish, and that 
will account for the fact that many non - technical 
ij words too were taken over, such as ai>, flg^ (juridical.?), 
' arrive (military.?), beast, change, cheer, cover, cry, debt 
(juridical.?), feeble, large, letter, manner, matter, nurse and 
nourish, place, point, price, reason, turn, use, and a great 
many other everyday words of very extensive employment. 
93. If, then, the English adopted so many French 
words because it was the fashion in every respect to 
imitate their 'betters', we are allowed to connect this 
adoption of non-technical words with that trait of their 
N character which in its exaggerated form has in modern 
times been termed snobbism or toadyism, and which 

1 Quoted by D. Behrens, Paul's Grundriss 1-963- 


V. The French. 

has made certain sections of the English people more 
interested in the births, deaths and especially marriages 
of dukes and marquises than in anything else outside 
their own small personal sphere. 

94. But when we trace this feature of snobbishness 
back to the first few centuries after the Norman conquest, 
we must not forget that there were great differences, so 
that some people would affect many French words and 
others would stick as far as possible to the native stock 
of words. We see this difference in the literary works 
that have come down to us. In Layamon's 'Brut', written 
very early in the thirteenth century and amounting in 
all to more than 56,000 short lines, the number of words 
of Anglo-French origin is only about 150.^ The *Orr- 
mulum', which was writter perhaps tw^enty years later, 
contains more than 20,000 lines, yet even Kluge, who 
criticizes the view that this very tedious work contains no 
French words, has not been able to find in it more than 
twenty odd words of French origin.^ But in the con- 
temporary prose work 'Ancrene Riwle', we find on 200 
pages about 500 French words. A couple of centuries 
later, it would be a much harder task to count the French 
words in any author, as so many words had already be- 
come part and parcel of the English language; but even 
then one author used many more than another. Chaucer 
undoubtedly employs a far greater number of French 
words than most other writers of his time. Nor would 
it be fair to ascribe all these borrowings to what I 
have mentioned as snobbism; the greater a writer's 
familiarity with French culture and literature, the 


1 Skeat, Principles of English Etymology , IT (1891) p. 8; 
Morris, Historical Outl. of Engl. Accidence (1885) p. 338. 

2 Kluge, Das franzosische element im Orrmulum, Englische 
Studien, XXII p. 179. 

Date of Adoption. q3 

greater would be his temptation to introduce French 
words for everything above the commonplaces of daily life. 
95. The following table shows the strength of the in- 
flux of French words at different periods; it comprises 
one thousand words (the first hundred French words in the 
New English Dictionary for each of the first nine letters 
and the first 50 for / and /) and gives the half-century to 
which the earliest quotation in that Dictionary belongs.^ 

Before 1050 .... 2 

105 1 — 1 100 2 

iioi — 1150 I 

1151 — 1200 15 

1201 — 1250 64 

1251 — 1300 127 

1301 — 1350 120 

1351 — 1400 180 

1401 — 1450 70 

145 1 — 1500 ^^ 

I50I— 1550 84 

I55I— 1600 91 

I60I — 1650 69 

1651 — 1700 34 

1701 — 1750 24 

1751 — 1800 16 

1801 — 1850 23 

1851 — 1900 2 


I I have followed the authority of the same Dictionary also in 
regard to the question of the origin of the words, reckoning thus as 
French some words which I should, perhaps, myself have called 
Latin. Derivative words that have certainly or probably arisen in 
English (e. g. daintily, damageable) have been excluded, as also 
those perfectly unimportant words for which the N. E. D. gives less 
than five quotations. Most of them cannot really be said to have 
ever belonged to the English language. Cf. also R. Mettig, Die 
franz. elemente im alt- und mittelengl. Engl. St. 41. i76fF.) 



QA V. The French. 

The list shows conclusively that the linguistic in- 
fluence did not begin immediately after the conquest, 
and that it was strongest in the years 125 1 — 1400, to 
which nearly half of the borrowings belong (42.7 p. c). 
Further it will be seen that the common assumption 
that the age of Dryden was particularly apt to intro- 
duce new words from French is very far from being correct. 

96. In a well-known passage, Robert of Gloucester 
(ab. 1300) speaks about the relation of the two lan- 
guages in England: 'Thus, he says, England came into 
Normandy's hand; and the Normans at that time iJ)o; 
it is important not to overlook this word) could speak 
only their own language, and spoke French just as they 
did at home, and had their children taught in the same 
manner, so that people of rank in this country who 
came of their blood all stick to the same language that 
they received of them, for if a man knows no French 
people will think little of him. But the lower classes 
still^ stick to English and to their own language. I 
imagine there are in all the world no countries that do 
not keep their own language except England alone. But 
it is well known that it is the best thing to know both 
languages, for the more a man knows the mpre is he 
worth.' This passage raises the question: How did com- 
mon people manage to learn so many foreign words? — 
and how far did they assimilate them? 

97. In a few cases the process of assimilation was 
facilitated by the fact that a French word happened to 
resemble an old native one; this was sometimes the 
natural consequence of French having in some previous 
period borrowed the corresponding word from some Ger- 

I yute 'yet'; sometimes curiously mistranslated, hold to their 
own £^ood speech. 


How was French learnt? nc 

manic dialect. Thus no one can tell exactly how much 
modern rich owes to OE. rice 'powerful, rich' and how 
much to French riche; the noun (Fr. and ME.) richesse 
(now riches) supplanted the early ME. richedom. The 
old native verb choose was supplemented with the noun 
choice from Fr. choix. OE. hergian and OFr. herier^ 
harier, run together in Mod. E. harry; OE. hege and Fr. 
haie run together in hay 'hedge, fence'. It is difficult 
to separate two main's, one of which is OE. mcegen 
'strength, might' and the other OFr. maine (Latin mag- 
nus] the root of both words is ultimately the same), cf. 
main sea and main force. The modern gain (noun and 
verb) was borrowed in the fifteenth century from French 
{gain, gaain; gagner gaaignier, cf. It. guadagnare, a Ger- 
manic loan), but it curiously coincided with an earlier 
noun gain (also spelt gein, geyn, gayne, etc., oldest form 
ga^henn), which meant 'advantage, use, avail, benefit, 
i remedy' and a verb gain (gayne, ge^^nenn) 'to be suit- 
able or useful, avail, serve', both from Old Norse. When 
French isle (now He) was adopted, it could not fail to 
remind the English of their old iegland. Hand and 
eventually it corrupted the spelling of the latter into is- 
land, Neveu (now spelled nephew) recalled OE. nefa^ 
meneye [menye, Fr. maisnie 'retinue, troop') recalled many 
(OE. menigeo), and lake, the old lacu 'stream, river. '^ 
There is some confusion between Eng. rest (repose) and 
OF. rest (remainder). In grammar, too, there were a few 
correspondences, as when nouns had the voiceless and 
the corresponding verbs the voiced consonants; French 
us — user, now use sb. pronounced [ju's], vb. [j^'z] just 
as Eng. house sb. [haus], vb. [hauz]; French grief — 
griever, Eng. grief — grieve just as half — halve. Note 
also the formation of nouns in -er [baker, etc.), which is 

I This is still the meaning of lake in some dialects. 



V. The French. 

hardly distinguishable from French formations in words 
like carpenter (Fr. -ier), interpreter (ME. interpretour, Fr. 
-eur), etc. But on the whole such more or less accidental 
similarities between the two languages were few in num- 
ber and could not materially assist the English popu- 
lation in learning the new words that were flooding their 

98. A greater assistance may perhaps have been de- 
rived from a habit which may have been common in con- 
versational speech, and which was at any rate not un- 
common in writing, that of using a French word side 
by side with its native synonym, the latter serving more 
or less openly as an interpretation of the former for the 
benefit of those who were not yet familiar with the more 
refined expression. Thus in the Ancrene Riwle (ab. 
1225): cherit6 )?et is luve (p. 8) | in desperaunce, pet is, 
in unhope & in unbileave forte beon iboruwen (p. 8) | 
Understonde^ )?et two manere temptaciuns — two kunne 
vondunges — beo^ (p. 180) | pacience, J^et is )?olemod- 
nesse (ibid.) | lecherie, pet is, golnesse (p. 198) | igno- 
raunce, pet is unwisdom & unwitenesse (p. 278). I quote 
from Behrens's collection of similar collocations^ the follow- 
ing instances that prove conclusively that the native 
word was then better known than the imported one: 
bigamie is unkinde [unnatural] J^ing, on engleis tale 
twiewifing (Genesis & Exod. 449) | twelfe iferan, pe 
Freinsce heo cleopeden dusze pers (Layamon I. I. 69) ] 
J7at craft: to lokie in J?an lufte, pe craft his ihote [is 
called] astronomic in o]7er kunnes speche [in a speech of a 
different kind] (ib. II. 2. 598). It is well worth observing 
that in all these cases the French words are perfectly 
familiar to a modern reader, while he will probably re- 

I Franz. Studien V. 2 p, 8 Cf. also 'of whiche tribe, that 
is to seye, kynrede Jesu Crist was bom' (Maundeville 67). 

Tautology. n ^ 

quire an explanation of the native words that served 
then to interpret the others. In Chaucer we find similar 
double expressions, but they are now introduced for a 
totally different purpose; the reader is evidently sup- 
posed to be equally familiar with both, and the writer uses 
them to heighten or strengthen the effect of the style^; 
for instance: He coude songes make and wel endyte (A 95) 
= Therto he coude endyte and make a thing (A 325) | 
faire and fetisly (A 124 and 273) | swinken with his handes 
and laboure (A 186) | Of studie took he most cure and 
most hede (A 303) | Poynaunt and sharp (A 352) | At 
sessiouns ther was he lord and sire (A 355). ^ In Caxton 
this has become quite a mannerism, see, e. g. I shal so 
awreke and avenge this trespace (Reynard 56, cf. p. 116 
advenge and wreke it) 1 in honour and worship (ib. p. 56) | 
olde and auncyent doctours (p. 62) | feblest and wekest 
(p. 64) I I toke a glasse or a mirrour (p. 83) | Now ye 

shal here of the mirrour \ the glas [P- 84) | good 

ne prof[yt (p, 86) | fowle and dishonestly (p. 94) | prouffyt 
and for dele (p. 103). It will be observed that with the 
exception of the last word, the language has preserved 
in all cases both the synonyms that Caxton uses side by 
side, so that we may consider this part of the English 
vocabulary as settled towards the end of the fifteenth 

99. Many of the French words, such as cry, claim, 
state, poor, change, and, indeed, most of the words enu- 

' 1 This use of two expressions for the same idea is extremely 
:[ common in the middle ages and the beginning of the modem 
5 period, and it is not confined to those cases where one was a 
f native and the other an imported word; see Kellner, Engl. 
\ Studien XX p. iiff. (1895); Greenough and Kittredge, Words 
j and their Ways, p. iisff.; so also in Danish, see ViHi. Andersen 
:| in Dania p. 86 ff. (1890) and Danske Studier 1893, p. jff. 
f 2 Cf. also, Curteys he was, lowly, and servisable (A 99); Cur- 
' teys he was, and lowly of servyse (A 250). 

Jespersen: English. 2nd ed. 7 

gg V, The French. 

merated above, (§ 82 — 92), and one might say, nearly 
all the words taken over before 1350 and not a few of 
those of later importation, have become part and parcel 
of the English language, so that they appear to us all 
just as English as the pre-Conquest stock of native words. 
But a great many others have never become so popular. 
There are a great many gradations between words of 
everyday use and such as are not at all understood by the 
common people, and to the latter class may sometimes 
belong words which literary people would think familiar 
to everybody. Hyde Clark relates an anecdote of a 
clergyman who blamed a brother preacher for using the 
word felicity J *I do not think all your hearers understood 
it; I should say happiness.' *I can hardly think,' said 
the other, 'that any one does not know what felicity 
means, and we will ask this ploughman near us. Come 
hither, my man ! you have been at church and heard 
the sermon; you heard me speak of felicity; do you know 
what it means?' *Ees, sir!' 'Well, what does felicity 
mean?' 'Summut in the inside of a pig, but I can't say 
altogether what.'^ — Note also the way in which Touch- 
stone addresses the rustic in As You Like It (V. I. 52) 
'Therefore, you Clowne, abandon, — which is in the 
vulgar leave, — the societie — which in the boorish is 
companie, — of this female, — which in the common is 
woman; which together is, abandon the society of this 
Female, or, Clowne, thou perishest; 01, to thy better 
understanding, dyest.' 

100. From what precedes we are now in a position 
to understand some at least of the differences that have 
developed in course of time between two synonyms when 
both have survived, one of them native, the other French. 


I A Grammar of the English Tongue. 4 th ed. London 1879. 


Synonyms. ng 

The former is always nearer the nation's heart than the 
latter, it has the strongest associations with everything 
primitive, fundamental, popular, while the French word 
is often more formal, more polite, more refined and has 
a less strong hold on the emotional side of Hfe. A cottage 
is finer than a hut, and fine people often live in a cottage, 
at any rate in summer 'The word hill was too vulgar 
and famihar to be apphed to a hawk, which had only a 
heak (the French term, whereas bill is the A. S. bile), 
'Ye shall say, this hauke has a large beke, or a short beke, 
and call it not bille' \ Book of St. Alban's, fol. a 6, back'.^ 
— To dress means to adorn, deck, etc., and thus generally 
presupposes a finer garment than the old word to clothe^ 
the wider signification of which it seems, however, to be 
more and more appropriating to itself. Amity means 
'friendly relations, especially of a public character be- 
tween states or individuals', and thus lacks the warmth of 
friendship. The difference between help and aid is thus 
indicated in the Funk-Wagnalls Dictionary: 'Help ex- 
presses greater dependence and deeper need than aid. 
In extremity we say 'God help me!' rather than 'God 
aid me !' In time of danger we cry 'help! help!' rather 
than ^aid! aid!' To aid is to second another's own exer- 
tions. We can speak of helping the helpless, but not of 
aiding them. Help includes aid, but aid may fall short 
of the meaning of help.' All this amounts to the same 
thing as saying that help is the natural expression, be- 
longing to the indispensable stock of words and there- 
fore possessing more copious and profounder associations 
than the more literary and accordingly colder word aid. 
Folk has to a great extent been superseded by people, 
chiefly, I suppose, on account of the political and social 
employment of the word; Shakespeare rarely uses folk 

I Skeat, The Works of G. Chaucer, vol. Ill p. 261. 


lOo ^ • The French. 

{4 times) and folks (ten times), and the word is evidently 
a low-class word with him; it is rare in the Authorized 
Version, and Milton never uses it; but in recent usage 
folk seems to have been gaining ground, partly, perhaps, 
from antiquarian and dialectal causes. Hearty and cor- 
dial made their appearance in the language at the same 
time (the oldest quotations 1380 and 1386, NED.), but 
where they signify the same thing their force is not the 
same, for *a hearty welcome' is warmer than 'a cordial 
welcome', and hearty has many applications that cordial 
has not (heartfelt, sincere; vigorous: a hearty slap on 
the back; abundant: a hearty meal), etc. Saint smacks 
of the official recognition by the Catholic Church, while 
holy refers much more to the mind. Matin(s) is used 
only with reference to church service, while morning is 
the ordinary word. Compare also darling with favourite, 
deep with profound, lonely with solitary, indeed with in 
fact, to give or to hand with to present or to deliver, love 
with charity, etc. 

101. In some cases the chief difference between the 
native word and the French synonym is that the former is 
more colloquial and the latter more literary, e. g. begin — 
commence, hide — conceal, feed — nourish, hinder — pre- 
vent, look for — search for, inner and outer — interior 
and exterior, and many others. In a few cases, however, 
the native word is more literary. Valley is the everyday 
word, and dale has only lately been introduced into the 
standard language from the dialects of the hilly northern 
counties. Action has practically supplanted deed in ordi- 
nary language, so that the latter can be reserved for 
more dignified speech. 

102. In spite of the intimate contact between French 
and English it sometimes happens that French words 
which have been introduced into other Germanic lan- 
guages and belong to their everyday vocabulary are not 

Colloquial and ' literary. , I O i 

found in English or are there much more felt to be foreign 
intruders than in German or Danish. This is true for 
instance of friseur, manchette, replique, of gene and the 
verb gener (the NED. has no instances of it, but a few 
are found in the Stanford Diet.). Serviette is rarer than 
napkin. Atelier is not common; it occurs in Thackeray's 
The Newcomes p. 242, where immediately afterwards 
the familiar word studio is used: did EngHsh artists go 
more to Italy and less to Paris to learn their craft than 
their Scandinavian and German confreres? To the same 
class belong the following words, which, when found in 
English books, are generally indicated to be foreign by 

; italic letters: na'ive^ bizarre, and motif, — the last word 

|| an interesting recent doublet of motive. 

103. As the grammatical systems of the two languages 
were very different, a few remarks must be made here 

I about the form in which French words were adopted. 
Substantives and adjectives^were nearly always taken 
over in the accusative case, which differed in most words 
from the nominative in having no s. The latter ending 
is, however, found in a few words, such as fitz (Fitzher- 
bert, etc.; in French, too, the nominative fils has ousted 
the old ace. fil', fitz is an Anglo-Norman spelling), fierce 
(0 Fr. nom. fiers, ace. fier), and James.^ In the plural, 

ll' Old French had a nominative without any ending and 
an accusative in -s, and English popular instinct natur- 

I But Chaucer Yizs by seint Jame (riming with name, D. 1443). 
A similar vacillation is found in the name Steven Stephen, 
where now the j-less form has prevailed, but where formerly 
the Fr. nom. was also found (seynt stevyns, Malory 104). — 
Where the French inflexion was irregular, owing to Latin 
stress shifting, etc., the accusative was adopted, in emperof 
(■our, O Fr. nom. emperere), companion (O Fr. nom. compain), 
neveu, nephew (O Fr. nom. nies) and others, but the nom. is 
kept in stre (O Fr. ace. seigno?), mayor (O Fr. maire, ace. 

IQ2 V. The French. 

ally associated the' 'latter form with the common 
English plural ending in -es. In course of time those 
words which had for a long time, in English as in 
French, formed their plural without any ending 
(e. -g. cas) were made to conform with the general rule 
(sg. case, pi. cases). ^ — French adjectives had the s 
added to them just like French nouns, and we find 
a few adjectives with the plural 5, as in the goddes 
celestials (Chaucer); letters patents survived as a fixed 
group till the time of Shakespeare (§ 85). But the 
general rule was to treat French adjectives exactly like 
English ones. 

104. As to the verbs, the rule is that the stem of the 
French present plural served as basis for the English 
form; thus (je survis), nous survivons, vous survivez, Us 
survivent became survive, (je resous), resolvons, etc., be- 
came resolve, O Fr. (je desjeun), nous disnons, etc., be- 
came dine; thus is explained the frequent ending -ish, in 
punish, finish, etc. English hound (to leap), accordingly, 
cannot be the French hondir, which would have yielded 
hondish, but is an English formation from the noun 
bound, which is the French bond. I think that levy is 
similarly formed on the noun levy, which is Fr. levee; but 
in sally the y represents the i which made the Fr. // mouilU. 
Where the French infinitive was imported it was generally 
in a substantival function, as in dinner, remainder, at- 
tainder, rejoinder, cf. the verbs dine, remain, attain, 
rejoin; so also the law terms merger, user, and misnomer. 
Still we have a few verbs in which the ending -er can 
hardly be anything else but the French infinitive ending: 
render (which is thereby kept distinct from rend), sur- 

I Note invoice, trace (part of a horse's harness), and quince, 
where the French plural ending now forms part of the English 
singular; cf. Fr. envoi, trait, coign. 

Grammar. 103 

render, tender (where the doublet tend also exists), and 
perhaps broider (embroider) . There is a curious parallel 
to the Norse bask and busk (79) in saunter, where the 
French reflective pronoun has become fixed as an insep- 
arable element of the word, from s'auntrer, another form 
for s'aventurer 'to adventure oneself. 

105. French words have, as a matter of course, parti- 
cipated in all the sound changes that have taken place 
in English since their adoption. Thus words with the 
long [i] sound have had it diphthongized into [ai], e. g. 
fine, price, lion. The long [u], written ou, has similarly- 
become [au], e. g. O Fr. espouse (Mod. Fr. Spouse), M. E. 
spouse, pronounced [spu'za], now pron. [spauz], Fr. tour, 
Mod. E. tower. Compare also the treatment of the 
vowels in grace, change, beast (OFr. beste), ease (Fr. aise)^ 
etc. Such changes of loan-words are seen everywhere: 
they are brought about gradually and insensibly. But 
there is another change which has often been supposed 
to have come about in a different manner. A great many 
words are now stressed on the first syllable which in 
French were stressed on the final syllable, and this is 
often ascribed to the inability of the English to imitate 
the French accentuation. All English words, it is said, 
had the stress on the first syllable, and this habit was 
unconsciously extended to foreign words on their first 
adoption into the language. We see this manner of 
treating foreign words in Icelandic at the present day. But 
the explanation does not hold good in our case. English 
had a few words with unstressed first syllable [be-, for-, 
etc., see above, § 25), and as a matter of fact, French 
words in English were for centuries accented in the 
French manner, as shown conclusively by Middle English 
poetry. It was only gradually that more and more 
words had their accent shifted on to its present place. 
The causes of this shifting were the same as are else- 

J04 ^- The French. 

where at work in the same direction.^ In many words 
the first syllable was felt as psychologically the most im- 
portant one, as in punish, finish, matter, manner, royal, 
army and other words ending with meaningless or form- 
ative syllables. The initial syllable very often received 
the accent of contrast. In modern speech we stress the 
otherwise unstressed syllables to bring out a contrast 
clearly, as in 'not oppose but suppose' or 'If on the one 
hand speech gives ^;vpression to ideas, on the other hand 
it receives impressions from them' (Romanes, Mental 
Evolution in Man, p. 238), and in the same manner we 
must imagine that in the days when real, formal, object, 
subject and a hundred similar words were normally stressed 
on the last syllable, they were so often contrasted with 
each other that the modern accentuation became grad- 
ually the habitual one. This will explain the accent of 
January, February, cavalry, infantry, primary, orient and 
other words. An equally powerful principle is rhythm, 
which tends to avoid two consecutive strong syllables; 
compare modern go down^stairs, but the ^downstairs 
room, St. Paul's church^yard, but the ^churchyard wall. 
Chaucer stresses many words in the French manner, 
except when they precede a stressed syllable, in which 
case the accent is shifted, thus co^syn (cousin), but ^cosyn 
^myn; in felici'te par^flt, but a ^verray ^parfit ^gentil ^knight; 
severe (secret), but in hecre wyse, etc. An instructive illus- 
tration is found in such a line as this (Cant. Tales 
D i486): 

In 'divers 'art and in di'vers fi'gures. 

These principles — value-stressing, contrast, rhythm — 
will explain all or most of the instances in which Eng- 
lish has shifted the French stress; but it is evident that 
it took a very long time before the new forms of the 

I See the detailed exposition in my Modern English Gram- 
mar (Heidelberg, Carl Winter) 1909 ch. V. 

Accent; Hybrids. 105 

words which arose at first only occasionally through their 
influence were powerful enough finally to supplant the 
older forms. ^ 

106. Not long after the intrusion of the first French 
words we begin to see the first traces of a phenomenon 
which was to attain very great proportions and which 
must now be termed one of the most prominent features 
of the language, namely hybridism. Strictly speaking, 
we have a hybrid (a composite word formed of elements 
from different languages) as soon as an English inflexion- 
al ending is added to a French word, as in the genitive 
the Duke's children or the superlative noblest, etc., and 
from such instances we rise by insensible gradations to 
others, in which the fusion is more surprising. From 
the very first we find verbal nouns in -ing or -ung formed 
from French verbs (indeed, they are found at a tim.e 
when they could not be formed from every native verb, 
§ 200), e. g. prechinge; riwlunge (Ancrene Riwle); scor- 
nunge and servinge (Layamon) ; spusinge (Owl & N.), 
Other instances of English endings added to French words 
are faintness (from the end of the fourteenth century), 
closeness (half a century later), secretness (Chaucer se- 
creenesse B 773), simpleness (Shakespeare and others), 
materialness (Ruskin), ahnormalness (Benson)^ etc. Fur- 
ther, a great many adjectives in -ly (courtly, princely, etc.) 
and, of course, innumerable adverbs with the same en- 
ding (faintly, easily, nobly); adjectives in -ful (beautiful, 
dutiful, powerful, artful) and -less (artless, colourless); 
nouns in -ship (courtship, companionship) and -dom 
(dukedom, martyrdom) and so forth. 

I In recent borrowings the accent is not shifted, of. 
machine, intrigue, where the retention of the French /-sound 
is another sign that the words are of comparatively modem 

lo6 V. The French. 

107. While hybrid words of this kind are found in 
comparatively great numbers in most languages, hybrids 
of the other kind, i. e. composed of a native stem and 
a foreign ending, are in most languages much rarer than 
in English. Before such hybrids could be formed, there 
must have been already in the language so great a num- 
ber of foreign words with the same ending that the form- 
ation would be felt to be perfectly transparent. Here 
are to be mentioned the numerous hybrids in -ess (shep- 
herdess, goddess; Wycliffe has dwelleresse; in a recent 
volume I have found 'seeress and prophetess'), in -ment 
(endearment and enlightenment are found from the 17th 
century, but bewilderment not before the 19th; wonder- 
ment, frequent in Thackeray; oddment, R. Kipling, hut- 
ment), in -age (mileage, acreage, leakage, shrinkage, 
wrappage, breakage, cleavage, roughage, shortage, etc.); 
in -ance (hindrance, used in the fifteenth century in the 
meaning 'injury'; in the signification now usual it is 
found as early as 1526, and perhaps we may infer from 
its occurring neither in the Bible, nor in Shakespeare, 
Milton, and Pope, that it was felt to be a bastard, though 
Locke, Cowper, Wordsworth, Shelley, and Tennyson ad- 
mit it; forbearance, originally a legal term; further- 
ance); in -ous (murderous; thunderous; slumberous is 
used by Keats and Carlyle); in -ry (fishery, bakery, etc.; 
gossipry, Mrs. Browning; Irishry; forgettery jocularly 
formed after memory); in -ty (oddity, womanity nonce- 
word after humanity): in -fy (fishify, Shakespeare; snug- 
gify, Ch, Lamb; Torify, Ch. Darwin; scarify, Fielding; 
tipsify, Thackeray; funkify; speechify^ with the corre- {^ 
sponding nouns in -fication (uglification, Shelley).'^ 

1 Cf. also 'Daphne — before she was happily treeified', Lowell, 
Fable for Critics. 

2 See below on hybrids with Latin and Greek endings 

(S 123). 


Hybridism. 1 07 

108. One of the most fertile English derivative endings 
is -able, which has been used in a great number of 
words besides those French ones which were taken 
over ready made (such as agreeable^ variable, tolerable). 
In comparatively few cases it is added to substantives 
(serviceable, companionable , marriageable, peaceable, 
seasonable). Its proper sphere of usefulness is in 
forming adjectives from verbs, rarely in an active 
sense [suitable = that suits, unshrinkable), but generally 
in a passive sense {bearable = that can or may be 
borne). Thus we have now drinkable, eatable, steer able 
(balloons) , weavable , unutterable , answerable , punish- 
able, unmistakable, etc., and hundreds of others, so 
that everybody has a feeling that he is free to form a 
new adjective of this kind as soon as there is any 
necessity for, or convenience in, using it, just as he 
feels no hesitation in adding -ing to any verb, new or 
old. And of course, no one ever objects to these ad- 
jectives (or the corresponding nouns in -ability) because 
they are hybrids or bastards, any more than one would 
object to forms like acting or remembering on the same 

109. These adjectives have now become so indispen- 
sable that the want is even felt of forming them from 
composite verbal expressions, such as get at. But though 
get-at-able and come-at-able are pretty frequently heard 
in conversation, most people shrink from writing or print- 
ing them. Sterne has come- at- ability. Smiles get-at- 
ability, and George Eliot in a letter knock-upable. Tenny- 
son, too, writes in a jocular letter, 'thinking of you as 
no longer the comeatable runupableto, smokeablewith 
J. S. of old.' Note here the place of the preposition in 
the last two adjectives, and compare 'enough to make 
the house unliveable in for a month' (The Idler, May 
1892, 366) and 'the husband being fairly good-natured 

Io8 V. The French. 

and livable-with' (Bernard Shaw, Ibsenism 41). It is 
obvious that these adjectives are too clumsy to be ever 
extensively used in serious writing. But there is an- 
other way out of the difficulty which is really much more 
conformable to the genius of the language, namely to 
leave out the preposition in all those cases where there 
can be no doubt of the preposition understood. Unac- 
countable {= that cannot be accounted for) has long 
been accepted by everybody; I have found it, for 
instance, in Congreve, Addison, Swift, Goldsmith, De 
Quincey, Miss Austen, Dickens and Hawthorne. 
Indispensable has been — well, indispensable, for two 
centuries and a half. Laughable is used by Shake- 
speare, Dryden, Carlyle, Thackeray, etc. Dependable, 
disposable , and available are in general use.^ All this 
being granted, it is difficult to see why reliable should 
be the most abused word of the English language. 
It is certainly formed in accordance with the funda- 
mental laws of the language; it is short and unam- 
biguous, and what more should be needed.? Those 
who measure a word by its age will be glad to hear 
that Miss Mabel Peacock has found it in a letter, 
bearing the date of 1624, from the pen of the Rev. 
Richard Mountagu, who eventually became a bishop. 
And those who do not like using a word unless it has 
been accepted by great writers will find a formidable 
array of the best names in Fitzedward Hall's list^ of 

1 Miss Austen writes, 'There will be work for five sum- 
mers before the place is liveable (Mansf. Park 216) = the 
above-mentioned liveable- in. Cf. below gazee and others in 
-ee (§ III) The principle of formation is the same as in 
Tvaite? 'he who waits on people', calle? 'he who calls on 
some one'. 

2 On English Adjectives in -able, with special reference to 
reliable. London 1877. Fitzedward Hall reverted to the sub- 
ject on several other occasions. 


Reliable. lOO 

authors who have used the word.^ It is curious to note 
that the word which is always extolled at the expense of 
reliable as an older and nobler word, namely trustworthy, 
is really much younger: at any rate, I have not been able 
to trace it further back than the beginning of the nine- 
teenth century; besides, any impartial judge will find 
its sound less agreeable to the ear on account of the 
consonant group — stw — and the heavy second 

110. Fitzedward Hall in speaking about the recent 
word aggressive^ says, 'It is not at all certain whether 
the French agressif suggested aggressive^ or was suggest- 
ed by it. They may have appeared independently of 
each other/ The same remark applies to a great many 
other formations on a French or Latin basis; even if the 
several components of a word are Romance, it by no 
means follows that the word was first used by a French- 
man. On the contrary, the greater facility and the greater 
boldness in forming new words and turns of expression 
which characterizes English generally in contradistinction 
to French, would in many cases speak in favour of the 
assumption that an innovation is due to an English mind. 
This I take to be true with regard to dalliance, which is 
i so frequent in ME. [dalyaunce, etc.) while it has not been 
recorded in French at all. The wide chasm between the 
most typical Enghsh meaning of sensible (a sensible 

1 Coleridge, Sir Robert Peel, John Stuart Mill, Abp. Long- 
ley, Samuel Wilberforce , Dickens, Charles Reade, Walter Bage- 
hot, Anthony Trollope, R. A. Proctor, Harriet Martineau, Car- 
dinal Newman, Gladstone, James Martineau, S. Baring-Gould, 
Sir G. O. Trevelyan, Sir Monier Williams, Sir Leslie Stephen, H. 
Maudsley, Saintsbury, Henry Sweet, Robinson Ellis, Thomas 
Arnold. In America, Washington Irving, Daniel Webster, Edw. 
Everett, G. P. Marsh; I leave out, rather arbitrarily I fear, six- 
teen of the names given by Fitzedward Hall. 

2 Modem English 314. 

I lo V. The French. 

man, a sensible proposal) and those meanings which it 
shares with French sensible and Lat. sensibilis, probably J 
shows that in the former meaning the word was an in- 
dependent English formation. Duration as used by- 
Chaucer may be a French word; it then went out of the 
language, and when it reappeared after the time of Shake- 
speare, it may just as well have been re-formed in Eng- 
land as borrowed; duratio does not seem to have existed 
in Latin. Intensitas is not a Latin word, and intensity is 
older than intensite. 

111. In not a few cases, the English soil has proved 
more fertilizing than the French soil from which words 
were transplanted. In French, for instance, mutin has 
fewer derivatives than in English, where we have mutine 
sb., mutine vb. (Shakespeare), mutinous, mutinously, 
mutinousness, mutiny sb., mutiny vb., mutineer sb., mu- 
tineer vb., mutinize, of which it is true that mutine and , 
mutinize are now extinct. We see the same thing in such 
a recent borrowing as clique, which stands alone in French ^ 
while in English two centuries have provided us with 
cliquedom, cliqueless, cliquery, cliquomania, cliqiiomaniac, 
clique, vb,, cliquish, cliquishness, cliquism, cliquy or cli- 
quey. From due we have duty, to which no French cor- 
respondent word has been found in France itself, although 
duete, duity, dewetS are found in Anglo-French writers; 
in English duty is found from the 13th century, and we 
have moreover duteous, dutiable, dutied, dutiful, dutifullyf 
dutifulness, dutiless, none of which appear to be older 
than the i6th century. Aim, the noun as well as the 
verb, is now among the most useful and indispensable 
words in the English vocabulary and it has some deriva- 
tives, such as aimer, aimful, and aimless, but in French 
the two verbs from which it originates, esmer < Lat. 
aestimare, and aasmer, < Lat. adaestimare, have totally .1 
disappeared. Note also the differentiations of the words 


English Formations. ill 

strange and estrange;'^ of entry (< Fr. entree^) and entrance^ 
while in French entrance has been given up; and the less 
perfect one of guaranty (action) and guarantee (person), 
not to speak of warrant and warranty. The extent to 
which foreign speech-material has been turned to 
account is really astonishing, as is seen, perhaps, most 
clearly in the extensive use of the derivative ending -ee. 
This was originally the French participial ending -e used 
in a very few cases such as apele^ E. appellee as opposed 
to apelor, E. appellor, nominee, presentee, etc. and then 
gradually extended in legal use to words in which such 
a formation would be prohibited in French by formal as 
well as syntactical reasons: vendee is the man to whom 
something is sold (I'homme ^ qui on a vendw quelque 
chose), cf. also referee, lessee, trustee, etc. Now, these 
formations are no longer restricted to juridical language, 
and in general literature there is some disposition to turn 
this ending to account as a convenient manner of forming 
passive nouns; Goldsmith and Richardson have lovee, 

Sterne speaks of 'the mortgager and mortgagee the 

jester and jestee'; further the gazee (De Quincey) = the 
one gazed at, staree (Edgeworth), cursee and laughee 
(Carlyle), flirtee, floggee, wishee, bargainee, beatee, examineCf 
callee (our callee = the man we call on), etc. Such a 
word as trusteeship is eminently characteristic of the 
composite character of the language: Scandinavian 
trust + a French ending used in a manner unparalleled 
!J in French -|- an old English ending. 

112, French influence has not been restricted to one 
particular period (see § 95), and it is interesting to com- 

1 Compare also the juridical estray and the ordinary stray, 
estate and state. 

2 This word has recently been re -adopted: entree 'made -dish 
served between the chief courses'. 

J I 2 V. The French. 

pare the forms of old loan-words with those of recent 
ones, in which we can recognize traces of the changes ^ 
the French language has undergone since medieval times. 
Where a ch in an originally French word is pronounced 
as in change, chaunt, etc. (with the sound-group tj), the y 
loan is an old one; where it is sounded as in champagne \ 
(with simple p, we have a recent loan. Chief is thus 
shown to belong to the first period, while its doublet 
chef (= chef de cuisine) is much more modern. It is 
curious that two petnames should now be spelled in the 
same way Charlie, although they are distinct in pro- 
nunciation: the masculine is derived from the old loan 
Charles and has, therefore, the sound [tJ], the feminine 
is from the recent loan Charlotte with [fj. Similar^ g as 
in giant and / as in jaundice [pronounced d^] are indicative 
of old loans, while the pronunciation [^] is only found in 
modern adoptions, such as rouge. Sometimes, however, 
recent loans are made to conform to the old practice; 
jaunty, gentle and genteel represent three layers of borrow- 
ing from the same word, but they have all of them the 
same initial sound. Other instances of the same French 
word appearing in more than one shape according to its 
age in English are saloon and salon, suit and suite, liquor 
and liqueur, rout 'big party, retreat' and route (the diph- 
thong in the former word is an English development of 
the long [u] § 105), quart, pronounced [kwo"t], and quart 
pronounced [ka"t] 'a sequence of four cards in piquet', 
of. also quarte or carte in fencing. 

113. In some cases, we witness a curious re-shaping 
of an early French loan-word, by which it is made more 
like the form into which the French has meanwhile de- 
veloped. This, of course, can only be explained by the 
uninterrupted contact between the two nations. Chaucer 
had viage just as Old French, but now the word is voyage; 
leal has given way to loyal; the noun flaute and the verb 

Early and Recent Loans. 1 1 7 

fioyten are now made into flute like mod. Fr. fluted Sim- 
ilarly the signification of ME. douten like that of OFr. 
douter was 'to fear' (cf. redoubt)^ but now in both lan- 
guages this signification has disappeared. Danger was at 
first adopted in the Old French sense of 'dominion, 
power', but the present meaning was developed in France 
before it came to England. The many parallelisms in the 
employment of cheer and Fr. chere could not very well 
have arisen independently in both languages at once. This 
continued contact constitutes a well-marked contrast be- 
tween the French and the Scandinavian influence, which 
seems to have been broken off somewhat abruptly after 
the Norman conquest. 

I Cf. below the Latinizing of many French words § 1 16. 


Jbsperskn: English. 2nd ed. g 

Chapter VL 

Latin and Greek. 

114. Although Latin has been read and written in ^ 
England from the Old English period till our own days, ' 
so that there has been an uninterrupted possibility of |j 
Latin influence on the English language, yet we may with 
comparative ease separate the latest stratum of loans 
from the two strata^ that we have already considered. It 
embodies especially abstract or scientific words, adopted 
exclusively through the medium of writing and never 
attaining to the same degree of popularity as words be- 
longing to the older strata. The words adopted are not 
all of Latin origin, there are perhaps more Greek than 
Latin elements in them, if we count the words in a big 
dictionary. Still the more important words are Latin, 
and most of the Greek words have entered into English 
through Latin, or have, at any rate, been Latinized m 
spelling and endings before being used in English, so that 
we have no occasion here to deal separately with the two 
stocks. The great historical event, without which thisd 
influence would never have assumed such gigantic di- 
mensions, was the revival of learning. Through Italy il| 
and France the Renaissance came to be felt in England 
as early as the fourteenth century, and since then the 
invasion of classical terms has never stopped, although 
the multitude of new words introduced was greater, 
perhaps, in the fourteenth, the sixteenth and the nine- 

The Renaissance. I j c 

teenth than in the intervening centuries. The same in- 
fluence is conspicuous in all European languages, but in 
English it has been stronger than in any other language, 
French perhaps excepted. This fact cannot, I think, be 
principally due to any greater zeal for classical learning 
on the part of the English than of other nations. The 
reason seems rather to be, that the natural power of 
resistance possessed by a Germanic tongue against these 
alien intruders had been already broken in the case of 
the English language by the wholesale importation of 
French words. They paved the way for the Latin words 
which resembled them in so many respects, and they had 
already created in English minds that predilection for 
foreign words which made them shrink from consciously 

I coining new words out of native material. If French 
words were more distingues than English ones, Latin 
words were still more so, for did not the French them- 
selves go to Latin to enrich their own vocabulary.? The 
first thing noticeable about this cJass of Latin importa- 

i tion is, therefore, that it cannot be definitely separated 
from the French loans. 

115. A great many words may with equal right be 
1 ascribed to French and to Latin, since their English form 
I would be the same in both cases and the first users would 
I probably know both languages. This is especially the 

case with those words which in French are not popular 
* survivals of spoken Latin words, but later borrowings 
t from literary Latin, mots savants, as Brachet termed them 
j in contradistinction to mots populaires. As examples of 
[words that may have been taken from either language, 
\ I shall mention only grave, gravity, consolation, solid, 
Hnfidel, infernal, position. 

116. A curious consequence of the Latin influence during 
jand after the Renaissance was that quite a number of 
'French words were remodelled into closer resemblance 



1 1 5 VI. Latin and Greek. 

with their Latin originals. Chaucer uses descrive (riming 
with on lyve 'alive' H. 121; still in Scotch), but in the 
1 6th century the form describe makes its appearance. 
Perfet and parfet (Fr. perfait, parfait) were the normal 
English forms for centuries. Milton writes perfeted 
(Areop. 10); but the c was introduced from the Latin, 
at first in spelling only, but afterwards in pronunciation 
as well.^ Similarly verdit has given way to verdict. Where 
Chaucer had peynture as in French (peinture), picture is 
now the established form. The Latin prefix ad is now 
seen in advice and adventure, while Middle English had 
avis [avys) and aventure. The latter form is still retained 
in the phrase at aventure, where however, a has been 
apprehended as the indefinite article (at a venture), and 
another remnant of the old form is disguised in saunter 
(Fr. s'aventurer *to adventure oneself). Avril (avrille) 
has been Latinized into April; and a modern reader doe s 
not easily recognize his February in ME. feouerele or 
feou£rrere^ (u = v, cf. fivrier). In debt and doubt, which 
used to be dette and doute as in French, the spelling only 
has been affected; compare also victuals for vittles (Fr. 
vitailles, cf. battle from bataille). Similarly bankerota [cf. 
Italian), banqueroute, bankrout (Shakesp.) had to give 
way to bankrupt; the oldest example of the p-form in the 
NED. dates from 1533. The form langage was used for 
centuries, before it became language by a curious crossing 
of French and Latin forms. Egal was for more than two 
centuries the commoner form; equal, now the only re- |l 
cognized form, was apparently a more learned form and J 
was used for instance in Chaucer's Astrolabe, while in his 
poems he writes egal; Shakespeare generally has equal, * 

1 Bacon writes {New Atlantis 15): all nations have enter- 
knowledge one of another. In recent similar words inter- is al- 
ways used. 

2 Juliana p. 78 , 79. , 


Remodelling of French Words. 117 

but egal is found a few times in some of the old editions 
of his plays. Tennyson tries to re-introduce egality by the 
side of equality, not as an ordinary word, however, but 
as applied to France specially (That cursed France with 
her egalities !' Aylmer's Field). French and Latin forms 
coexist, more or less differentiated, in complaisance and 
complacence (complacency), genie (rare) and genius, base 
and basis (Greek). Certainty (Fr.) and certitude (Lat.) are 
often used indiscriminately, but there is now a tendency 
to restrict the latter to merely subjective certainty, as in 
Cardinal Newman's 'my argument is: that certitude was 
a habit of mind, that certainty was a quality of proposi- 
tions; that probabilities which did not reach to logical 
certainty, might suffice for a mental certitude', etc.^ — 
Note also the curious difference made between critic with 
stress on the first syllable, adjective^ and nomen agentis 
(from Lat., or Greek direct.? or through French.?) and 
critique with stress on the second syllable, nomen actionis 
(late borrowing from Fr.); Pope uses critick'd as a parti- 
ciple (stress on the first), while a verb critique with stress 
on the last syllable is found in recent use ; criticize, which 
since Milton has been the usual verb, is a pseudo-Greek 

117. Intricate relations between French and Latin are 
sometimes shown in derivatives: colour is from French, 
as is evident from the vowel in the first syllable [a]; but 
in discoloration the second syllable is sometimes made 
[kol] as from Latin, and sometimes (kAl] as from French. 
Compare also example from French, exemplary from Latin. 
Machine with machinist and machinery are from the 
French, witness the pronunciation [me'ji'n]; but machinate 

1 Apologia pro Vita sua. New impression, London 1900 
p. 20. 

2 With the by -form criticaL 

1 1 8 VI. Latin and Greek. 

and machination are taken direct from Latin and accord- 
ingly pronounced [maekineit, maeki'neijan]; so these two 
groups which ought by nature to belong together are I 
kept apart, and no one knows whether the adjective 
machinal should go with one or the other group, some 
dictionaries pronouncing [m9'j'i'n9l] and others ['maekinal] 
— a suggestive symptom of the highly artificial state of 
the language ! 

118. It would be idle to attempt to indicate the number 
of Latin and Greek words in the English language, as 
each new treatise on a scientific subject adds to their 
number. But it is interesting to see what proportion of 
the Latin vocabulary has passed into English. Professors 
J. B. Greenough and G. L. Kittredge have counted the 
words beginning with A in Harper's Latin Dictionary, 
excluding proper names, doublets, parts of verbs, and 
adverbs in -e and -ter. 'Of the three thousand words there - 
catalogued, one hundred and fifty-four (or about one in 
twenty) have been adopted bodily into our language in 
some Latin form, and a little over five hundred have 
some English representative taken, or supposed to be 
taken, through the French. Thus we have in the English 
vocabulary about one in four or five of all the words 
found in the Latin lexicon under A. There is no reason 
to suppose that this proportion would not hold good 
approximately for the whole alphabet.'^ 

119. It must not be imagined that all the Latin words 
as used in English conform exactly with the rules of 
Latin pronunciation or with the exact classical meanings. 
'My instructor, says Fitzedward HalP, took me to task \\ 

1 Words mia their Ways, 1902, p. 106. 

2 Fitzedward Hall, Two Trifles. Printed for the Author 
1895. I have changed his symbol for stress, indicating here i^ 
as elsewhere the beginning of the strong syllable by a 
prefixed 1 . 

Deviations from Latin. 


for saying ^doctrinal. 'Where an English word is from 
Latin or Greek, you should always remember the stress 
in the original, and the quantity of the vowels there.' 
I replied: 'If others, in their solicitude to pro^pagate re- 
finement, choose to be irritated or ^excited, because of 
what they take to be my genuine ignorance in oratory, 
they should at least be sure that their discomposure is 
not gratuitous.' — Among words used in English with 
a different signification from the classical one, may be 
mentioned enormous (Latin enormis 'irregular', in English 
formerly also enorm and enormious), item (Latin item 'also', 
used to introduce each article in a list, except the first), 
ponder (Lat. ponderare 'to weigh, examine, judge', transi- 
tive), premises ('adjuncts of a building', originally things 
set forth or mentioned in the beginning), climax (Greek 
klimax 'a ladder or gradation'; in the popular sense of 
culminating point it is found in Emerson, Dean Stanley, 
John Morley, Miss Mitford and other writers of repute), 
bathos (Greek bathos 'depth'; in the sense of 'ludicrous 
descent from the elevated to the commonplace' it is due 
to Pope; the adjective bathetic, wrongly formed on the 
analogy of pathetic, was first used by Coleridge). It 
should be remembered, however, that when once a certain 
pronunciation or signification has been firmly established 
in a language, the word fulfils its purpose in spite of ever 
so many might-have-beens, and that, at any rate, cor- 
rectness in one language should not be measured by the 
yard of another language. Transpire is perfectly legiti- 
mate in the sense 'to be emitted through the pores of the 
skin' and in the derived sense 'to become known, to be- 
come public gradually' although there is no Latin verb 
transpirare in either of these senses; if, therefore, the 
modern journalistic use of the verb in the sense of 'happen' 
('a terrible murder has again transpired in Whitechapel') 
is objectionable, it is not on account of any deviation 

I20 VI. Latin and Greek. 

from Latin usage, but because it has arisen through a 
vulgar misunderstanding of the English signification of 
an EngHsh word. Stuart Mill exaggerates the danger of 
such innovations, when he writes: 'Vulgarisms, which 
creep in nobody knows how, are daily depriving the 
English language of valuable modes of expressing thought. 

To take a present instance: the verb transpire 

Of late a practice has commenced of employing this 
word, for the sake of finery, as a mere synonym of to 
happen: 'the events which have transpired in the Crimea^ 
meaning the incidents of the war. This vile specimen 
of bad English is already seen in the despatches of noble- 
men and viceroys: and the time is apparently not far 
distant when nobody will understand the word if used in 

its proper sense The use of 'aggravating' for 

'provoking', in my boyhood a vulgarism of the nursery^ 
has crept into almost all newspapers, and into many ^ 
books; and when writers on criminal law speak of aggra- 
vating and extenuating circumstances, their meaning, 
it is probable, is already misunderstood.'^ Let me add 
two small notes to Mill's remarks. First, that aggravate 
in the sense of 'exasperate, provoke' is exemplified in the 
NED. from Cotgrave (i6ii), T. Herbert (1634), Richard- 
son (1748) — thus some time before Mill heard it in his 
nursery — and Thackeray (1848). And secondly, that 
the verb which Mill uses to explain it, provoke, is here 
used in a specifically English sense which is nearly as far 
removed from the classical signification as that of ag- 
gravate is. But we shall presently see that the English 
have taken even greater liberties with the classical il 

120. When the influx of classical words began, it had 
its raison d'etre in the new world of old, but forgotten 

I Stuart Mill, A System of Logic, People's edition, 1886^ 
P- 451. 

Ideas and Words. I 2 i 

ideas, then first revealed to medieval Europe. Instead 
of their narrow circle of everyday monotonousness, 
people began to suspect new vistas, in art as well as in 
science, and classical literature became a fruitful source 
of information and inspiration. No wonder then, that 
scores and hundreds of words should be adopted together 
with the ideas they stood for, and should seem to the 
adopters indispensable means of enriching a language 
which to them appeared poor and infertile as compared 
with the rich storehouses of Latin and Greek. But as 
times wore on, the ideas derived from classical authors 
were no longer sufficient for the civilized world, and, just 
as it will happen with children outgrowing their garments, 
the modern mind outgrew classicism, without anybody 
noticing exactly when or how. New ideas and new habits 
of life developed and demanded linguistic expression, and 
now the curious thing happened that classical studies had 
so leavened the minds of the educated classes that even 
when they passed the bounds of the ancient world they 
drew upon the Latin and Greek vocabulary in preference 
to their own native stock of words. 

181. This is seen very extensively in the nomenclature 
of modern science, in which hundreds of chemical, bo- 
tanical, biological and other terms have been framed 
from Latin and Greek roots, most of them compound 
words and some extremely long compounds. It is cer- 
tainly superfluous here to give instances of such forma- 
tions, as a glance at any page of a comprehensive dic- 
tionary will supply a sufficient number of them, and as 
one needs only a smattering of science to be acquaint- 
ed with technical words from Latin and Greek that 
would have struck Demosthenes and Cicero as bold, 
many of them even as indefensible or incomprehensible 
innovations. It is not, perhaps, so well known that quite 
a number of words that belong to the vocabulary of ord- 

J 22 VI. Latin and Greek. 

inary life and that are generally supposed to have the 
best-ascertained classical pedigree, have really been 
coined in recent times more or less exactly on classical 
analogies. Some of them have arisen independently in 
several European countries. Such modern coinages are, 
for instance, eventual with eventuality, immoral, fragmental 
and fragmentary, primal, annexation, fixation and affixa- 
tion, climatic. There are scores of modern formations in 
-ism^, e. g. absenteeism, alienism, classicism, colloquialism, 
favouritism, individualism, mannerism, realism, not to 
speak of those made from proper names, such as Swin- 
burnism, Zolaism, etc. Among the innumerable words 
of recent formation in -ist may be mentioned dentist, 
economist, florist, jurist, oculist, copyist (formerly copist 
as in some continental languages), determinist, economist, 
ventriloquist, individualist, plagiarist, positivist, socialist, 
terrorist, nihilist, tourist. For calculist the only author 
quoted in the NED. is Carlyle. Scientist has often been 
branded as an 'ignoble Americanism' or 'a cheap and H 
vulgar product of trans-Atlantic slang', but Fitzedward 
Hall has pointed out that it was fabricated and advocated, 
in 1840, together with physicist, by Dr. Whewell. Who- 
ever objects to such words as scientist on the plea that 
they are not correct Latin formations, would have to 
blot out of his vocabulary such well-established words 
as suicide, telegram, botany, sociology, tractarian, vege- 
tarian, facsimile and orthopedic; but then, happily, people 
are not consistent. 

122. Authors sometimes coin quasi-classic words \\ 
without finding anybody to pass them on, as when Mil- 
ton writes 'our inquisiturient Bishops' (Areop. 13). Cole- ,, 
ridge speaks of 'logodcedaly or verbal legerdemain', ' 

I See Fitzedward Hall, Modem English, p. 311. His lists 
have also been utilized in the rest of this paragraph. id 

Innovations. ^ i2;\ 

Thackeray of a lady's 'viduous mansion' (Newc. 794), 
Dickens of 'vocular exclamations' (Oliver Twist) ; Tennyson 
writes in a letter (Life I. 254) 'you range no higher in my 
andrometer' ; Bulwer-Lytton says *a cat the most viparious 
[meaning evidently 'tenacious of life'] is limited to nine 
lives'; and Mrs. Humphrey Ward 'his air of old-fashioned 
punctilium.'^ I have here on purpose mixed correct 
and incorrect forms, jocular and serious words, because 
my point was to illustrate the love found in most English 
writers of everything Latin or Greek, however unusual 
or fanciful. Sometimes jocular 'classicisms' survive and 
are adopted into everybody's language, such as omnium 
gatherum (whence Thackeray's bold heading of a chapter 
'Snobbium Gatherum'), circumbendibus (Goldsmith, Cole- 
ridge) and tandem, which originated in a University pun 
on the two senses of English 'at length'. 

123. Hybrids, in which one of the component part was 
French and the other native English, have been men- 
tioned above (§ 106 f.). Here we shall give some examples 
of the corresponding phenomenon with Latin and Greek 
elements, some of which may, however, have been im- 
ported through French. The ending -ation is found in 
starvation, backwardation, and others; note also the Amer- 
ican thiinderation ('It was an accident, sir.' 'Accident 
the thunderation', Opie Read, Toothpick Tales, Chicago 
1892, p. 35). Johnsoniana, Miltoniana, etc., are quite 
modern; the ending ana alone is now also used as a de- 
tached noun. In -ist we have the American walkist, 
which is interesting as denoting a professional walker 
and therefore distinguished by the more learned ending. 
Compare also turfite and the numerous words in -ite 

I Dictionaries recognize punctilio, a curious transformation 
of Spanish puntillo; there is a late Latin punctillum , but not 
with the meaning of 'punctiliousness'. 


VI. Latin and Greek. 

derived from proper names: Irvingite, Ruskinite, etc. 
The same ending is frequently used in mineralogy and 
chemistry, one of the latest additons to these formations 
being fumelessite = smokeless gunpowder. Hybrids in 
-ism (cf. § 12 1) abound; heathenism has been used by 
Bacon, Milton, Addison, Freeman and others; witticism 
was first used by Dryden, who asks pardon for this new 
word ; hlock-headism is found in Ruskin ; further funnyism, 
free-lovism, etc.; the curious wegotism may be classed with 
the jocular drinkitite on the analogy of appetite. Girlicide, 
2iiter suicide, is another jocular formation (Smedley, Frank 
Fairlegh I 190, not in NED.). To the same sphere belong 
Byron's weatherology and some words in -ocracy^ such as 
landocracy, shopocracy, barristerocracy, squattocracy and 
G. Meredith's snipocracy (Evan Harrington 174, from snip 
ai a nickname for a tailor). On the other hand squirearchy 
(with squirearchical) seems to have quite established it- 
self in serious language. Among verbal formations must 
be mentioned those in -ize: he womanized his language 
(Meredith, Egoist 32), Londonizing (ibd. 80), soberize, etc. 
Adjectives are formed in -ative: talkative, babblative, 
scribblative, and soothative, of which only the first is rec- 
ognized; in -aceous: gossipaceous (Darwin, Life and Let- 
ters I 375), in -arious: burglarious (Stevenson, Dynamiter 
130), and -iacal: dandiacal (Carlyle, Sartor 188). Even if 
many of these words are 'nonce - words', it cannot 
be denied that the process is genuinely English and 
perfectly legitimate — within reasonable limits at any 

124. Some Latin and Greek prepositions have in re- 
cent times been extensively used to form new words. 
Ex-, as in ex-king, ex-head-master, etc.^, seems first to 
have been used in French, but it is now common to most 

I 'A pair of ex -white satin shoes' (Thackeray). 

Hybrids. I 2 5 

or all Germanic languages as well; in English this form- 
ation did not become popular till little more than a 
century ago. Anti-: the anti-taxation movement; an anti- 
foreign party; 'Mr. Anti-slavery Clarkson' (De Quincey, 
Opium- Eater 197); 'chairs unpleasant to sit in — anti- 
caller chairs they might be named' (H. Spencer, Facts 
and Comments 85). Co-: 'a friend of mine, co-godfather 
to Dickens's child with me' (Tennyson, Life II 114); 
'Wallace, the co-formulator of the Darwinian theory' 
(Clodd, Pioneers of Evolution 68). De-, especially with 
verbs in -ize: de-anglicize, de-democratize, deprovincialize, 
denationalize; less frequently as in de-tenant, de-miracle 
(Tennyson). Inter-: intermingle, intermix, intermarriage, 
interbreed, inter-communicate, inter-dependence, etc. 
International was coined by Bentham in 1780; it marks 
linguistically the first beginning of the era when relations 
between nations came to be considered like relations 
between citizens, capable of peaceful arrangement according 
to right rather than according to might. A great 
many other similar adjectives have since been formed: 
intercollegiate, interracial, interparliamentary, etc. Where 
no adjective existed, the substantive is used unchanged, 
but the combination is virtually an adjective: interstate 
affairs; ^.n inter-island sted^mev ; 'international, inter-club, 
inter-team, inter-college or inter-school contests' (quoted 
in NED.). Pre-: the pre-Darwinian explanations; pre- 
nuptial friendships (Pinero, Second Mrs. Tanqueray, p. 6, 
what are called on p. 8 'ante-nuptial acquaintances'); 
'in the pre-railroad, pre-telegraphic period' (G. Eliot); 
the pre-railway city; the pre-board school; a bunch of 
pre- Johannesburg Transvaals; the pre-mechanical civil- 
ized state (all these are quotations from H. G. Wells) ; in 
your pre-smoking days (Barrie). Pro-: the pro-Boers; 
pro-foreign proclivities; a pro-Belgian, or rather pro-King 
Leopold speaker. As any number of such derivatives or 

126 ^I« Latin and Greek. 

compounds can be formed with the greatest facihty, the 
utility and convenience of these certainly not classical 
expedients cannot be reasonably denied, though it may 
be questioned whether it would not have been better to 
utilize English prepositions for the same purposes, as 
is done with after- (an after-dinner speech) and sometimes 
with before- ('the before Alfred remains of our language', 
Sweet; 'smoking his bef ore-breakfast pipe', Conan Doyle). 
A few words must be added on re- which is used in a 
similar manner in any number of free compounds, such 
as rebirth, and especially verbs: re -organize, re-sterilize, 
re-submit, re-pocket, re-leather, re-case etc. Here re- 
is always strongly stressed and pronounced with a long 
vowel [i*], and by that means these recent words are 
in the spoken language easily distinguished from the 
older set of r^-words, where re is either weakly stressed 
or else pronounced with short [e]. We have therefore 
such pairs as recollect = to remember, and re-collect = to 
collect again; he recovered the lost umbrella and had it 
re-covered; reform and re-form (reformation and re-form- 
ation), recreate and re-create, remark and re-mark, 
resign and re-sign, resound and re-sound, resort and re-sort. 
In the written language the distinction is not always 

125. Latin has influenced English not only in vocabu- 
lary, but also in style and syntax. The absolute participle 
(as in 'everything considered', or 'this being the case') 
was introduced at a very early period in imitation of 
the Latin construction.^ It is comparatively rare in Old 
English, where it occurs chiefly in close translations from 
Latin. In the first period of Middle English it is equally 

I Morgan Callaway, The Absolute Participle in Anglo-Saxon. 
Baltimore 1889. - Charles Hunter Ross , The Absolute Participle 
in Middle and Modern English, Baltimore 1893. 

Syntax. 127 

rare, but in the second period it becomes a little more 
frequent. Chaucer seems to have used it chiefly in imi- 
tation of the Italian construction, but this Italian in- 
fluence died out with him, and French influence did very 
little to increase the frequency of the construction. In 
the beginning of the Modern English period the absolute 
participle, though occurring more often than formerly, 
'had not become thoroughly naturahzed. It limited it- 
self to certain favourite authors where the classical element 
largely predominated, and was used but sparingly by 
authors whose style was essentially English/ (Ross, 
p. 38.) But after 1660, when English prose style devel- 
oped a new phase, which was saturated with classical 
elements, the construction rapidly gained ground and 
was finally fixed and naturalized in the language. There 
are some other Latin idioms which authors tried to imi- 
tate, but which have always been felt as unnatural, so that 
now they have been dropped, for instance who for he who 
or those who as in 'sleeping found by whom they dread' 
(Milton, P. L. I. 1333), further such interrogative and 
relative constructions as those found in the following 
quotations. 'To do what service am I sent for hither?' 
(Shakesp., R 2 IV. i. 176) and 'a right noble and pious 

lord, who had he not sacrificed his life we had 

not now mist and bewayl'd a worthy patron' (Milton, 
Areop. 51). 

126. Latin grammar was the only grammar taught in 
those days, and the only grammar found worthy of 
study and imitation. 'That highly discipHned syntax 
which Milton favoured from the first, and to which he 
tended more and more, was in fact, the classical syntax, 
or, to be more exact, an adaptation of the syntax of 
the Latin tongue,' says D. Masson^ and when he adds, 

I Poetical Works of Milton, 1890, vol. Ill, p. 74—5- 

J 28 ^^' Latin and Greek. 

*It could hardly fail to be so Even now, questions 

in English syntax are often settled best practically, if a 
settlement is wanted, by a reference to Latin construc- 
tion', he expresses a totally erroneous conception which 
has been, and is, unfortunately too common, although 
very little linguistic culture would seem to be needed to 
expose its fallacy. Nowhere, perhaps, has this miscon- 
ception been more strongly expressed than in Dryden's 
preface to 'Troilus and Cressida', where he writes: 'How 
barbarously we yet write and speak your Lordship knows, 
and I am sufficiently sensible in my own English. For 
I am often put to a stand in considering whether what 
I write be the idiom of the tongue, or false grammar 
and nonsense couched beneath that specious name of 
Anglicism, and have no other way to clear my doubts 
but by translating my English into Latin, and thereby 
trying what sense the words will bear in a more stable 
language.' I am afraid that Dryden would never have 
become the famous writer he is, had he employed this 
practice as often as he would have us imagine. But it 
was certainly in deference to Latin syntax that in the 
later editions of his Essay on Dramatic Poesy he changed 
such phrases as 'I cannot think so contemptibly of the 
age I live in' to 'the age in which I live'; he speaks some- 
where^ of the preposition at the end of the sentence as 
a common fault with Ben Jonson 'and which I have but 
lately observed in my own writings.' The construction 
Dryden here reprehends is not a 'fault' and is not con- 
fined to Ben Jonson, but is a genuine English idiom of 
long standing in the language and found very frequently 
in all writers of natural prose and verse. The omission 
of the relative pronoun, which Dr. Johnson terms 'a 

I I quote this second-hand, see J. Earle, Efiglish Prose 267; 
Hales, Notes to Milton's Areopagitica , p. 103. 

Syntax and Style. 129 

colloquial barbarism' and which is found only seven or 
eight times in all the writings of Milton, and (according 
to Thum) only twice in the whole of Macaulay's History, 
abounds in the writings of such authors as Shakespeare, 
Bunyan, Swift, Fielding, Goldsmith, Sterne, Byron, 
Shelley, Dickens, Thackeray, Tennyson, Ruskin, etc., 
etc. In Addison's well-known 'Humble Petition of Who 
and Which'^ these two pronouns complain of the injury 
done to them by the recent extension of the use of that. 
'We are descended of ancient Families, and kept up our 
Dignity and Honour many Years till the Jacksprat that 
supplanted us.' Addison here turns all historical truth 
topsy-turvy, for that is much older as a relative pronoun 
than either who or which] but the real reason of his pre- 
dilection for the latter two was certainly their conform- 
ance to Latin relative pronouns, and there can be no 
doubt that his article, assisted by English grammars and 
the teaching given in schoolrooms, has contributed very 
much to restricting the use of that as a relative pronoun 
■ — in writing at least. Addison himself, when editing the 
Spectator in book-form, corrected many a natural that to 
a less natural who or which, 

127. As to the more general effect of classical studies 
on English style, I am very much inclined to think that 
Darwin and Huxley are right as against most school- 
masters. 'Ch. Darwin had the strongest disbelief in the 
common idea that a classical scholar must write good 
English; indeed he thought that the contrary was the 
case. '2 Huxley wrote to the Times, Aug. 5, 1890:^ 'My 
impression has been that the Genius of the English lan- 
guage is widely different from that of Latin; and that 

1 The Spectator, no. 78, May 30, 1711. 

2 Life and Letters of Ch. Darwin, 1887, I p. 155. 

3 Quoted by J. Earle, English Prose, 487. 

Jbspersen: English. 2Qd ed. < 

J 90 - VI, Latin and Greek. 

the worst and the most debased kinds of English style 
are those which ape Latinity. I know of no purer English 
prose than that of John Bunyan and Daniel Defoe; I 
doubt if the music of Keats's verse has ever been sur- 
passed; it has not been my fortune to hear any orator 
who approached the powerful simplicity, the Hmpid sin- 
cerity, of the speech of John Bright. Yet Latin Hterature 
and these masters of Enghsh had little to do with one 
another.' As 'in diesem bund der dritte' might be men- 
tioned Herbert Spencer, who expressed himself strongly 
to the same effect in his last book.^ 

128. To return to the vocabulary. We may now con- 
sider the question: Is the Latin element on the whole 
beneficial to the English tongue or would it have been 
better if the free adoption of words from the classical 
languages had been kept within much narrower limits? 
A perfectly impartial decision is not easy, but it is hoped 
that the following may be considered a fair statement 
of the most important pros and cons. The first advan- 
tage that strikes the observer is the enormous addition 
to the English vocabulary. If the English boast that 
their language is richer than any other, and that their 
dictionaries contain a far greater number of words than j 
German and French ones, the chief reason is, of course, 
the greater number of foreign and especially of French 
and Latjn words adopted. 'I trade,' says Dryden, 'both 
with the living and the dead for the enrichment of our 
native language.' 

129. But this wealth of words has its seamy side too. 
The real psychological wealth is wealth of ideas, not of 
mere names. 'We have more words than notions, half a 
dozen words for the same thing', says Selden (Table Talk'' 
LXXVI). Words are not material things that can be 

I Facts and Comments, 1902, p. 70. 


Wealth of Words. 


heaped up like money or stores of food and clothes, 
from which you may at any time take what you want. 
A word to be yours must be learnt by you, and possessing 
it means reproducing it. Both the process of learning 
and that of reproducing it involve labour on your part. 
Some words are easy to handle, and others difficult. 
The number of words at your disposal in a given language 
is, therefore, not the only thing of importance; their 
quality, too, is to be considered, and especially the ease 
with which they can be associated with the ideas they 
are to symbolize and with other words. Now many of 
the Latin words are deficient in that respect, and this 
entails other drawbacks to speakers of English, as will 
presently appear. 

ISO. It will be argued in favour of the classical ele- 
ments t hat many of them fill up gaps in the native st ock 
of words, so that they serve to express ideas which would 
have been nameless but for them. To this it may be 
objected that the resources of the original language 
should not be underrated. In most, perhaps in all cases, 
it would have been possible to find an adequate ex- 
pression in the vernacular or to coin one. The tendency 
to such economy in Old English and the ease with which 
felicitous terms for new ideas were then framed by 
means of native speech-material, have been mentioned 
above. But Httle by little English speakers lost the habit 
of looking first to their own language and utilizing it 
to the utmost before going abroad for new expressions. 
People who had had their whole education in Latin and 
[[ had thought all their best thoughts in that language to 
\[ an extent which is not easy for us moderns to realize, 
often found it easier to write on abstract or learned 
;j subjects in Latin than in their own vernacular, and when 
ii they tried to write on these things in English, Latin 
[! words would constantly come first to their minds. Mental 


J 2 VI. Latin and Greek. 

laziness and regard to their own momentary convehience 
therefore led them to retain the Latin word and give 
it only an English termination. Little did they care for 
the convenience of their readers, if they should happen 
to be ignorant of the classics, or for that of unborn 
generations, whom they forced by their disregard for 
their own language to carry on the burden of committing 
to memory words and expressions which were really 
foreign to their idiom. If they have not actually dried 
up the natural sources of speech — for these run on as 
fresh as ever — yet they have accustomed their country- 
men to cross the stream in search of water, to borrow an 
expressive Danish locution. 

131. There is one class of words which seems to be 
rather sparingly represented in the native vocabulary, so 
that classical formations are extremely often resorted to, 
namely the adjectives. It is, in fact, surprising how 
many pairs we have of native nouns and foreign adjec- 
tives, e. g. mouth: oral; nose: nasal] eye: ocular; mind: 
mental; son: filial; ox: bovine; worm: vermicular; house: 
domestic; the middle ages: medieval; book: literary; moon: 
lunar; sun: solar; star: stellar; town: urban; man: human, 
virile, etc., etc. In the same category we may class such 
pairs as money: monetary, pecuniary; letter: epistolary; 
school: scholastic, as the nouns, though originally for- 
eign, are now for all practical purposes to be considered 
native. We may note here English proper names and 
their Latinized adjectives, e. g. Dorset: Dorsetian; Ox- 
ford: Oxonian; Cambridge: Cantabrigian; Gladstone: 
Gladstonian. Lancaster has even two adjectives, Lan- 
castrian (in medieval history) and Lancasterian (schools, 
Joseph Lancaster, 177 1 — 1838). It cannot be pretended 
that all these adjectives are used on account of any 
real deficiency in the English language, as it has quite 
a number of endings by which to turn substantives into 

Adjectives. 1^3 

adjectives: -en (silken), -y (flowery), -ish (girlish), -ly 
(fatherly), -like (fishlike), -some (burdensome), -ful (sin- 
ful), and these might easily have been utilized still more 
than they actually have been. In point of fact, we possess 
not a few native adjectives by the side of more learned 
ones, e. g. fatherly: paternal; motherly: maternal; brotherly: 
fraternal (but only sisterly, as sororal is so rare as to be 
left out of account) ; further watery: aquatic or aqueous; 
heavenly: celestial; earthy, earthly, earthen: terrestrial; 
timely: temporal; daily: diurnal; truthful: veracious; etc. 
In some cases the meanings of these have become more 
or less differentiated, the EngHsh words having often 
lost an abstract sense which they formerly had and which 
might have been retained with advantage. If the word 
sanguinary is now extensively used it is due to the 
curious twisting of the meaning of bloody in vulgar speech 
(cf. 244). Kingly, royal, and regal have now sHghtly 
different appHcations, but as royal in French, kongelig 
in Danish, and koniglich in German cover them all, 
English might have been content with one word instead 
of three, 

132. Besides, in a great many cases it is really con- 
trary to the genius of the language to use an adjective 
at all. Where Romance and Slavonic languages very 
often prefer a combination of a noun and an adjective the 
Germanic languages combine the two ideas into a com- 
pound noun. Birthday is much more English than natal 
day (which is used, for instance, in Wordsworth's 75th 
Sonnet), and eyeball than ocular globe, but physiologists 
think it more dignified to speak of the gustatory nerve 
than of the taste nerve and will even say mental nerve 
(Lat. mentum 'chin') instead of chin nerve in spite of 
the unavoidable confusion with the familiar adjective 
mental. Mere position before another noun is really the 
most English way of turning a noun into an adjective, 

17 A VI. Latin and Greek. 

e. g. the London market, a Wessex man, Yorkshire pud- 
ding, a strong Edinburgh accent, a Japan table, Venice 
glasses, the Chaucer Society, the Droeshout picture, a 
Gladstone bag, imitation Astrakhan, 'Every tiger madness 
muzzled, every serpent passion kill'd' (Tennyson).^ It is 
worth noting that the English adjective corresponding to 
family is not familiar, v^hich. has been somewhat estranged 
from its kindred, but family: family reasons, family affairs, 
family questions, etc. The unnaturalness of forming Latin 
adjectives is, perhaps, also shown by the vacillation often 
found between different endings, as in feudatary and 
feudatory, festal and festive. From labyrinth no less than 
six adjectives have been found: labyrinthal, labyrinthean, 
labyrinthian, labyrinthic, labyrinthical and labyrinthine. 
Many adjectives are quite superfluous; Shakespeare 
never used either autumnal, hibernal, vernal, or estival, 
and he probably never missed them. Instead of 
hodiernal and hesternal we have luckily other expressions 
(to-day's post; the questions of the day; yesterday's 
news). Most of us can certainly do without gressorial 
(birds), avuncular (a favourite with Thackeray: 'Clive, 
in the avuncular gig'; 'the avuncular banking house'; 
'the avuncular quarrel', all from The Newcomes), 
osculatory (processes = kissing; ib.) , lachrymatory (he 
is great in the 1. line; ib.) , aquiline ('What! am I an 
eagle too.? I have no aquiline prentensions at all', 
ib.)2 — and a great many similarly purposeless ad- 

133. More than in anything else the richness of the 
English language manifests itself in its great number of 

1 Shakespeare did not scruple to write 'the Carthage queen', 
'Rome gates', 'Tiber banks', even ' through faire Verona streets'. 
Cf. below, § 210. 

2 Thus used in a different manner from the familiar aqui- 
line nose. 

1 Synonyms. 1 35 

synonyms, whether we take this word in its strict sense 
of words of exactly the same meaning or in the looser 

I sense of words with nearly the same meaning. It is 
evident that the latter class must be the most valuable 

I as it allows speakers to express subtle shades of thought. 
Juvenile does not signify the same thing as youthful, pon- 
derous as weighty, portion as share, miserable as wretched. 
Legible means 'that can be read', readable generally 'worth 
reading'. Sometimes the Latin word is used in a more 
limited, special or precise sense than the English, as is 
seen by a comparison of identical and same, science and 
knowledge, sentence and saying, latent or occult and hidden. 
Breath can hardly now be called a synonym of spirit ('The 
spirit does not mean the breath', Tennyson), and simi- 
larly edify, which is still used by Spenser in the con- 
crete sense of 'building up', is now used exclusively 
with a spiritual signification, which its former synonym 
build can never have. Homicide is the learned, abstract, 
colourless word, while murder denotes only one kind of 
manslaughter, and killing is the everyday word with a 
much vaguer signification (being applicable also to ani- 
mals); there is a very apposite quotation from Coleridge 
in the NED.: '(He) is acquitted of murder — the act 
was manslaughter only, or it was justifiable homicide*. 
The learned word magnitude is more specialized than 
greatness or size (which is now thoroughly English, but is 
a very recent development of assize in a curiously modi- 
fied sense). The Latin masculine is more abstract than 
the English manly, which generally implies an emotional 
element of praise, the French male has not exactly the 
same import as either, and the Latin virile represents 
a fourth shade, while for the other sex we have feminine, 
womanly and womanish, the differences between which 
are not parallel to those between the first series of 

J 95 ^I- Latin and Greek. 

134. These examples will suffice to illustrate the syn- 
onymic relations between classical and other words. 
It will be seen that it is not always easy to draw a line 
or to determine exactly the different shades of meaning 
attached to each word; indeed, a comparison of the 
definitions given in various essays on synonyms and in 
dictionaries, and especially a comparison of these de- 
finitions with the use as actually found in various writ- 
ers, will show that it is in many cases a hopeless task 
to assign definite spheres of signification to these words. 
Sometimes the only real difference is that one term is 
preferred in certain collocations and another in others. 
Still, it is indubitable that very often the existence of 
a double or triple assortment of expressions will allow 
a writer to express his thoughts with the greatest pre- 
cision imaginable. But on the other hand, only those 
whose thoughts are accurate and well disciplined attain 
to the highest degree of Hnguistic precision, and the use 
in speech and writing of the same set of words by loose 
and inexact thinkers will always tend to blur out any 
sharp lines of demarcation that may exist between such 
synonymous terms as do not belong to their every-day 
stock of language. 

135. However, even where there is no real difference 
in the value of two words or where the difference is 
momentarily disregarded, their existence may not be en- 
tirely worthless, as it enables an author to avoid a tri- 
vial repetition of the same word, and variety of expres- 
sions is generally considered one of the felicities of style. 
We very often see English authors use a native and a 
borrowed word side by side simply, it would seem, to 
amplify the expression, without modifying its meaning. 
Thus 'of blind forgetjulnesse and dark oblivion' (Shake- 
speare, in Buckingham's strongly rhetorical speech, R 3 
III. 7. 129). The manifold multiform flower' (Swinburne, 

Synonyms. I^y 

Songs bef. Sunr. lo6). A perfectly natural variation of 
three expressions is seen in: 'the Bushman story is just 
the sort of story we expect from Bushmen, whereas the 
Hesiodic story is not at all the kind of tale we look for 
from Greeks'. (A. Lang, Custom and Myth 54.) Further 
examples: 'I went upstairs with my candle directly. It 
appeared to my childish fancy, as I ascended to the bed- 
room ' 'He asked me if it would suit my con- 
venience to have the light put out; and on my answering 
'yes', instantly extinguished it'. 'The phantom slowly 
approached. When it came near him, Scrooge bent down' ; 
'they are exactly unlike. They are utterly dissimilar in 
all respects' (all these from Dickens). 'We who boast 
of our land of freedom, we who live in the country of 
liberty.' 'I could not repress a half smile as he said this; 
a similar demi-manifestation of feeling appeared at the 
same moment on Hunsden's lips.' This kind of variation 
evidently does not always lead to the highest excellence 
of style. I quote from Minto^ Samuel Johnson's com- 
parison between punch and conversation: 'The spirit, 
volatile and fiery, is the proper emblem of vivacity and 
wit; the acidity of the lemon will very aptly figure pun- 
gency of raillery and acrimony of censure; sugar is the 
natural representative of luscious adulation and gentle 
complaisance; and water is the proper hieroglyphic of 
easy prattle, innocent and tasteless.' This is not far 
from Mr. Micawber's pihng up of words ('to the best 

of my knowledge, information, and belief to wit, 

in manner following, that is to say'), which gives Dickens 
the occasion for the following outburst: 

'In the taking of legal oaths, for instance, deponents 
seem to enjoy themselves mightily when they come to 
several good words in succession, for the expression of 

I Manual of English Prose Literature, 3rd ed. 1896, p. 418. 


\1. Latin and Greek. 

one idea; as, that they utterly detest, abominate, and i 
abjure, or so forth; and the old anathemas were made 
relishing on the same principle. We talk about the i 
tyranny of words, but we like to tyrannize over them J 
too; we are fond of having a large superfluous estabhsh- | 
ment of words to wait upon us on great occasions; we 
think it looks important, and sounds well. As we are 
not particular about the meanings of our liveries on 
state occasions, if they be but fine and numerous enough, 
so the meaning or necessity of our words is a second- 
ary consideration if there be but a great parade of them. 
And as individuals get into trouble by making too great 
a show of liveries, or as slaves when they are too numer- 
ous rise against their masters, so I think I could mention 
a nation that has got into many great difficulties, and 
will get into many greater, from maintaining too large 
a retinue of words.' [David Copperfield, p. 702.)^ ;^>j 

136. No doubt many of the synonymous terms intro- 
duced from Latin and Greek had best been let alone. No 
one would have missed pharos by the side of lighthouse, 
or nigritude by the side of blackness. The native words 
cold, cool, chill, chilly, icy, frosty might have seemed suf- 
ficient for all purposes, without any necessity for im- 
porting frigid, gelid, and algid, which, as a matter of fact, 
are neither found in Shakespeare nor in the Authorized 
Version of the Bible nor in the poetical works of Milton, 
Pope, Cowper, and Shelley. 

137. Apart from the advantage of being able con- 
stantly to make a choice between words possessing a 
different number of syllables and often also presenting 
a difference in the place of the accent, poets will often 

I Mr. Micawber also has the following delightful piece of 
bathos: 'It is not an avocation of a remunerative description — 
in other words, it does ?tot pay.' 


Big Words. l^g 

find the sonorous Latin words better for their purposes 
than the short native ones. In some kinds of prose 
writing, too, they are felt to heighten the tone, and add 
dignity, even majesty, to the structure of the sentence. 
The chief reason of this seems to be that the long word 
takes up more time. Instead of hurrying the reader or 
listener on to the next idea, it allows his mind to dwell 
for a longer time upon the same idea; it gives time for 
his reflexion to be deeper and especially for his emotion 
to be stronger. This seems to me more important than 
the two other reasons given by H. Spencer (Essays, II, 
p. 14) that 'a voluminous, mouth-filling epithet is, by 
its very size, suggestive of largeness or strength' and that 
'a word of several syllables admits of more emphatic 
articulation (.?); and as emphatic articulation is a sign 
of emotion, the unusual impressiveness of the thing 
named is implied by it.' Let me quote here also a 
quaint passage (not to be taken too seriously) from 
Howell (New English Grammar, 1662, p. 40): 'The 
Spanish abound and delight in words of many syllables, 
and where the English expresseth himself in one 
syllable, he doth in 5 or 6, as thoughts pensamientos, 
fray levantamiento &c, which is held a part of wisdom, 
for while they speak they take time to consider of the 

138. It is often said that the classical elements are 
commendable on the score of international intelligibility, 
and it is certain that many of them, even of those formed 
during the last century on more or less exact Latin 
and Greek analogy, are used in many other civilized 
countries as well as in England. The utility of this is 
evident in our days of easy communication between the 
nations; but on the whole its utility should not be valued 
beyond measure. If the thing to be named is one of 
everyday importance, national convenience should cer- 

j^o VI- Latin and Greek. 

tainly be considered before international ease; there- 
fore to wire and a wire are preferable to telegraph and 
telegram.'^ Scientific nomenclature is to a great extent 
universal, and there is no reason why each nation should 
have its own name for foraminifera or monocotyledones. But 
so much of science is now becoming more and more the 
property of everybody and influences daily life so deeply 
that the endeavour should rather be to have popular 
than learned names for whatever in science is not in- 
tended exclusively for the specialist. Sleeplessness is a 
better name than insomnia, and foreigners who know 
English enough to read a medical treatise in it will be no 
more perplexed by the word than an Englishman read- 
ing German is by schlajlosigkeit. Foreign phoneticians 
have had no difficulty in understanding Melville Bell's 
excellent nomenclature and have even to a great extent 
adopted the English terms of front, mixed, hack, etc. in 
preference to the more cumbersome palatal, gutturo- 
palatal, and guttural. It is a pity that half-vowel (Googe 
1577) and half-vowelish (Ben Jonson) should have been 
superseded by semi-vowel and semi-vowel-like. Among 
English words that have been in recent times adopted 
by many foreign languages may be mentioned cheque, 
box (in a bank), trust, film (in photography), sport, jockey, 
sulky, gig, handicap, dock, waterproof, tender, coke (Ger- 
man and Danish koks or sometimes with Pseudo- English 
spelling coaks), so that even to obtain international cur- 
rency a word need not have a learned appearance or be 
derived from Greek and Latin roots. Besides, many 
of the latter class are not quite so international as might 
be supposed, as their English significations are unknown 
on the continent (pathos, physic, concurrent, competition. 

I And why not use wireless as a verb too? 'Admiral N. 
has wirelessed that a Russian man-of-war is in sight', etc. 

Internationality. iai 

actual, eventual, injury); sometimes, also, the ending 
is different , as in principle (Fr. principe , etc.) , in- 
dividual (Fr. individu, German individuum), chemistry 
(chimie , chemie) , botany (botanique) , fanaticism (fana- 

139. It is possible to point out a certain number of 
inherent deficiencies which affect parts of the vocabulary 
borrowed from the classical language. Mention has al- 
ready been made (§ 26) of the stress-shifting which is so 
contrary to the general spirit of Germanic tongues and 
which obscures the relation between connected words, 
especially in a language where unstressed syllables are 
generally pronounced with such indistinct vowel sounds as 
in English. Compare, for instance, solid and solidity, 
pathos and pathetic, pathology and pathologic, pacify and 
pacific (note that the first two syllables of pacification, 
where the strongest stress is on the fourth syllable, va- 
cillate between the two corresponding pronunciations). 
The incongruity is especially disagreeable when native 
names are distorted by means of a learned derivative 
ending, as when Milton has the stress shifted on to the 
second syllable and the vowel changed (in two different 
ways) in Miltonic and Miltonian; cf. also Baconian, 
Dickensian, Taylorian, Spenserian, Canadian, Dorset- 
tan, etc. 

140. Another drawback is shown in the relation be- 
tween emit and immit, emerge and immerge. While in 
Latin emitto and immitto, emergo and immergo were easily 
kept apart, because the vowels were distinct and double 
consonants were rigorously pronounced double and so 
kept apart from single ones, the natural English pronun- 
ciation will confound them, just as it confounds the first 
syllables of immediate and emotion. Now, as the meaning 
of e- is the exact opposite of in-, the two pairs do not 
go well together in the same language. The same is true of 

J 4.2 ^I- Latin and Greek. 

illusion and elusion.'^ A still greater drawback arises 
from the two meanings of initial in, which is sometimes 
the negative prefix and sometimes the preposition. Ac- 
cording to dictionaries investigahle means (i) that may 
be investigated, (2) incapable of being investigated, and 
'infusible (i) that may be infused or poured in, (2) in- 
capable of being fused or melted. Importable, which is 
now only used as derived from import (capable of being 
imported) had formerly also the meaning 'unbearable', 
and improvable similarly had the meaning of 'incapable 
of being proved' though it only retains that of 'capable 
of being improved'. What Shakespeare in one passage 
(Temp. II. I. 37) expresses in accordance with modern 
usage by the word uninhabitable he elsewhere calls in- 
habitable (Even to the frozen ridges of the Alpes, Or any 
other ground inhabitable, R 2 I. i. 65), and the ambi- 
guity of the latter word has now led to the curious re- 
sult that the positive adjective corresponding to inhabit 
is habitable and the negative uninhabitable. The first 
syllable of inebriety is the preposition in-, so that it means 
the same thing as the rare ebriety 'drunkenness', but Th. 
Hook mistook it for the negative prefix and so, subtract- 
ing in-, made ebriety mean 'sobriety'. ^ Illustrious is 
used in Shakespeare's Cymb. I. 6. 109 as the negative 
of lustrous, while elsewhere it has the exactly opposite 
signification. Fortunately this ambiguity is Hmited to 
a comparative small portion of the vocabulary.^ 

1 Illiterate spellers will often write illicit for elicit, enumerable 
for innumerable, etc. Many words have had, and some still have, 
two spellings, with e7i- (em-) from the French, and with in- (im-) 
from the Latin {enquire, inquire, etc.) jH 

2 See quotation in Davies, Supplementary English Glossary "^ 

3 If invaluable means generally 'very valuable' and some- J 
times 'valueless', the case is obviously different from the above, '-i 

Want of Harmony. 1 43 

141. Loan-words do not necessarily n^ake a language 
inharmonious. In Finnish, for instance, in spite of 
numerous loans from a variety of languages, the prevailing 
impression is one of unity, apart perhaps from some of 
the most recent Swedish words. The foreign elements 
have been so assimilated in sound and inflexion as to 
be recognizable as foreign only to the eye of a philologist. 
The same may be said of the pre-Conquest borrowings 
from Latin into English, of the Scandinavian and of the 
most important among the French loans, nay even of a 
great many recent loans from exotic languages. Wine 
and tea, bacon and eggs, orange and sugar, plunder and 
war, prison and judge — all are not only indispensable, 
but harmonious elements of English. But while most 
people are astonished on first hearing that such words 
have not always belonged to their language, no philo- 
logical training is required to discover that phenomenon or 
diphtheriaor intellectual on latitudinarian3.rQ out of harmony 
with the real core or central part of the language. Every 
j one must feel the incongruity of such sets of words as 
I father — paternal — parricide or of the abnormal plurals 
which break the beautiful regularity of nearly all English 
! substantives — phenomena, nuclei, larvce, chrysalides, indices^ 
etc. The occasional occurrence of such blundering plurals as 
animalcules and ignorami is an unconscious protest against 
the prevalent pedantry of schoolmasters in this respect.^ 

I ' He may also see giraffes, lions or rhinoceros. The mention 
of this last word reminds me of a problem, which has tormented 
me all the time that I have been in East Africa, namely, what 
is the plural of rhinoceros? The conversational abbreviations, 
'rhino', 'rhinos', seem beneath the dignity of literature, and 
to use the sporting idiom by which the singular is always put 
for the plural is merely to avoid the difficulty. Liddell and 
Scott seem to authorize 'rhinocerotes' which is pedantic, but 

, 'rhinoceroses' is not euphonious.' Sir Charles Eliot, The East 

I, Africa Protectorate (1905) P- 266. 

jAA VI. Latin and Greek. 

142. The unnatural state into which the language has ^ 
been thrown by the wholesale adoption of learned words 

is further manifested by the fact that not a few of them 
have no fixed pronunciation; they are, in fact, eye-words 
that do not really exist in the language. Educated people 
freely write them and understand them when they see 
them written, but are more or less puzzled when they have 
to pronounce them. Dr. Murray relates how he was once 
present at a meeting of a learned society, where in the 
course of discussion he heard the word gaseous systema- 
tically pronounced in six different ways by as many emin- 
ent physicists. (NED., Preface.) Diatribist is by Murray 
and the Century Dictionary stressed on the first, by 
Webster on the second syllable, and the same hesitation 
is found with phonotypy, photochromy, and many similar 
words. This is, however, beaten by two so well-known 
words as hegemony and phthisis, for each of which diction- 
aries record no less than nine possible pronunciations 
without being able to tell us which of these is the preval- 
ent or preferable one. I doubt very much whether anal- 
ogous waverings can be found in any other language. 

143. The worst thing, however, that can be said 
against the words that are occupying us here is their 
difficulty and the undemocratic character which is a 
natural outcome of their difficulty. A great many of 
them will never be used or understood by anybody that 
has not had a classical education^. There are usually no 
associations of ideas between them and the ordinary 
stock of words, and no likenesses in root or in the form- 
ative elements to assist the memory. We have here 

I Sometimes they are not even understood by the erudite 
themselves, Gestic in Goldsmiths 'skill'd in gestic lore' (Trav- 
eller 253) is taken in all dictionaries as meaning 'legendary, 
historical' as \{ ixon\ gest , OYx. geste 'sX.Qixy , romance'; but the 
context shows conclusively that 'pertaining to bodily movement, 
esp. dancing' (NED.) must be the meaning; cf, Lat. gestus 


Malapropisms. 14c 

none of those invisible threads that knit words together 
in the human mind. Their great number in the language 
is therefore apt to form or rather to accentuate class 
divisions, so that a man's culture is largely judged of 
by the extent to which he is able correctly to handle 
these hard words in speech and in writing — certainly 
not the highest imaginable standard of a man's worth. 
No literature in the world abounds as English does in 
characters made ridiculous to the reader by the manner 
in which they misapply or distort 'big' words. Shake- 
speare's Dogberry and Mrs. Quickly, Fielding's Mrs. Slip- 
slop, Smollet's Winifred Jenkins, Sheridan's Mrs. Mala- 
prop, Dickens's Weller senior, Shillaber's Mrs. Parting- 
ton, and footmen and labourers innumerable made fun 
of in novels and comedies might all of them appear in 
court as witnesses for the plaintiff in a law-suit brought 
against the educated classes of England for wilfully mak- 
ing the language more complicated than necessary and 
thereby hindering the spread of education among all 
classes of the population. 

144. Different authors vary very greatly with regard 
to the extent to which they make use of such 'choice 
words, and measured phrase above the reach of ordinary 
men'. So much is said on this head in easily accessible 
textbooks on literature that I need not repeat it here. 
Unfortunately the statistical calculations given there of 
the percentage of native and of foreign words in different 
writers are not quite to the point, for while they generally 
include Scandinavian loans among native words, they 
reckon together all words of classical origin, although 
such popular words as cry or crown have evidently quite 

'gesture'. Arista? chy has been wrongly interpreted in most 
dictionaries as 'a body of good men in power', while it is 
derived from the proper name Aristarch and means 'a body 
of severe critics'. (Fitzedward Hall, Modern English 143.) 

Jbspersen: English. 2nd ed. 10 


VI. Latin and Greek. 

a different standing in the language from learned words 
like auditory or hymenoptera. The culmination with regard 
to the use of learned words in ordinary literary style was 
reached in the time of Dr. Samuel Johnson. I can find 
no better example to illustrate the effect of extreme 
'Johnsonese' than the following: — 

'The proverbial oracles of our parsimonious ancestors 
have informed us, that the fatal waste of our fortune is 
by small expenses, by the profusion of sums too little 
singly to alarm our caution, and which we never suffer 
ourselves to consider together. Of the same kind is the J 
prodigahty of life; he that hopes to look back hereafter 
with satisfaction upon past years, must learn to know 
the present value of single minutes, and endeavour to 
let no particle of time fall useless to the ground.'^ 

145. In his Essay on Madame D'Arblay Macaulay 
gives some delightful samples of this style as developed 
by that ardent admirer of Dr. Johnson. Sheridan refused 
to permit his lovely wife to sing in public, and was 
warmly praised on this account by Johnson. 'The last 
of men,' says Madame D'Arblay, 'was Doctor Johnson 
to have abetted squandering the delicacy of integrity 
by nullifying the labours of talent.' To be starved to 
death is 'to sink from inanition into nonentity.' Sir 
Isaac Newton is 'the developer of the skies in their em.- 
bodied movements', and Mrs. Thrale, when a party of 
clever people sat silent, is said to have been 'provoked 
by the dulness of a taciturnity that, in the midst of such 
renowned interloculors, produced as narcotic a torpor as 
could have been caused by a death the most barren of 

I Minto {Manual of Engl. Prose Lit 422) translates this as 
follows: 'Take care of the pennies', says the thrifty old proverb, 
• and the pounds will take care of themselves.' In like manner 
we might say, 'Take care of the minutes, and the years will 
take care of themselves.' 

Johnsonese, jaj 

all human faculties/ (Macaulay, Essays, Tauchn. ed. V. 

p. 65.) 

146. In the nineteenth century a most happy reaction 
set in in favor of 'Saxon' words and natural expressions ; 
and it is highly significant that Tennyson, for instance, 
prides himself on having in the 'Idylls of the King' used 
Latin words more sparingly than any other poet. But 
still the malady lingers on, especially with the half-edu- 
cated. I quote from a newspaper the following story: 
The young lady home from school was explaining. 'Take 
an egg', she said, 'and make a perforation in the base 
and a corresponding one in the apex. Then apply the 
lips to the aperture, and by forcibly inhaling the breath 
the shell is entirely discharged of its contents.' An old 
lady who was listening exclaimed: 'It beats all how folks 
do things nowadays. When I was a gal they made a 
hole in each end and sucked.' — To a different class 
belongs that master of Saxon English, Charles Lamb, 
who begins his 'Chapter on Ears' in the following way: 
*I have no ear. Mistake me not, reader, — nor imagine 
that I am by nature destitute of those exterior twin appen- 
dages, hanging ornaments, and (architecturally speaking) 
handsome volutes to the human capital. Better my 
mother had never borne me. I am, I think, rather 
delicately than copiously provided with those conduits; 
and I feel no disposition to envy the mule for his 
plenty, or the mole for her exactness , in those latjiyrin- 
thine inlets — those indispensable side - intelligencers.' 
0. W. Holmes, in his 'Our Hundred Days in Europe' 
avoids the simple expression 'a shaving machine' and 
'beard', and writes instead 'a reaping machine which 
gathered the capillary harvest of the past twenty - four 

lours in short, a lawn-mower for the mas- 

,:uline growth of which the proprietor wishes to rid 
•lis countenance.' 


148 VI. Latin and Greek. 

- 147. Of course, the authors of these two'sample) aim 
in them at a certain humorous effect, and very often 
similar circumlocutions are consciously resorted to in 
conversation to obtain a ludicrous effect, as 'he ampu- 
tated his mahogany' (cut his stick, went off), 'to agitate j 
the communicator' (ring the bell), 'are your corporeal 
functions in a condition of solubility.?*', *a sanguinary 
nasal protuberance', 'the Recent Incision' (the New Cut, 
a street in London), 'the Grove of the Evangehst' (St. 
John's Wood in London), etc. When Mr. Bob Sawyer 
asked 'I say, old boy, where do you hang out.?' Mr. Pick 
wick replied that he was at present suspended at the 
George and Vulture. (Dickens, Pickw. II 13.) Punch 
somewhere gives the following paraphrases of well-known 
proverbs: 'Iniquitous intercourses contaminate proper 
habits. In the absence of the feline race, the mice give 
themselves up to various pastimes. Casualties w411 take 
place in the most excellently conducted family circles. 
More confectioners than are absolutely necessary are apt 
to ruin the potage.' (Quoted in Fitzgerald's Miscellanies^ 
p. 166). Similarly *A rolling stone gathers no moss' is 
paraphrased 'Cryptogamous concretion never grows On 
mineral fragments that decline repose'. Some Latin and 
Greek words will scarcely ever be used except in jocular 
or ironical speech, such as sapient (wise), histrion (actor),j 
a virgin aunt (maiden aunt), hylactism (barking), edacious^ 
(greedy), the genus Homo (mankind), etc. 

148. But how many words are there not which belong 
virtually to the same class, but are used in dead earnest 
by people who know that many big words are found in 
the best authors and who want to show off their education 
by avoiding plain everyday expressions and by couching 
their thoughts in a would-be refined style.I* When Canning 
wrote the inscription graven on Pitt's monument in the 
London Guildhall, an Alderman felt much disgust at 



the grand phrase, 'he died poor', and wished to substitute 
'he expired in indigent circumstances'. Mr. Kington 
Oliphant, who relates this (The New English II 232), 
justly remarks, 'Could the difference between the 
scholarlike and the vulgar be more happily marked .>' 
James Russell Lowell , in the Introduction to the 
Second Series of his Biglow Papers, has a list of what 
he calls the old and the new styles of newspaper 
writing, which I find so characteristic that I select a few 
samples : — 

New Style. 

A vast concourse was as- 
sembled to witness. 

Old Style. 
A great crowd came to see. 

Great fire. 

The fire spread. 

Man fell. 

Sent for the doctor 

Began his answer. 
He died. 

Disastrous conflagration. 

The conflagration extended 
its devastating career. 

Individual was precipitated. 

Called into requisition the 
services of the family phy- 

Commenced his rejoinder. 

He deceased, he passed out 
of existence, his spirit 
quitted its earthly habit- 
ation, winged its way to 
eternity, shook off its bur- 
den, etc. 

149. I do not deny that somewhat parallel instances 
of stilted language might be culled from the daily press 
of most other nations, but nowhere else are they found 
in such plenty as in English, and no other language lends 
itself by its very structure to such vile stylistic tricks as 
English does. Wordsworth writes: 'And sitting on the 
grass partook The fragrant beverage drawn from 

I CO VI. Latin and Creek. 

China's herb', to which Tennyson remarked : 'Why could 
he not have said 'And sitting on the grass had tea?'^ 
Gissing in one of his novels says of a clergyman: 'One 
might have suspected that he had made a list of uncom- 
mon words wherewith to adorn his discourse, for certain 
of these frequently recurred. 'Nullifidian', 'morbific', 're- 
nascent', were among his favourites. Once or twice he 
spoke of 'psychogenesis', with an emphatic enunciation 
which seemed to invite respectful wonder'. ^ And did not 
little Thomas Babington Macaulay, when four years old, ^ 
reply to a lady who took pity on him after he had spilt 
some hot coffee over his legs, 'Thank you, madam, the 
agony is abated'? And does not a language which pos- 
sesses, besides the natural expression for each thing, two 
or three sonorous equivalents, tempt a writer into what 
Lecky hits off so well when he says of Gladstone: 'He 
seemed sometimes to be labouring to show with how 
many words a simple thought could be expressed or 

150. To sum up: the classical words adopted since 
the Renaissance have enriched the English language very 
greatly and have especially increased its number of syn- 
onyms. But it is not every 'enrichment' that is an 
advantage, and this one comprises much that is really 
superfluous, or worse than superfluous, and has, more- 
over, stunted the growth of native formations. The inter- 
national currency of many words is not a full compen- 
sation for their want of harmony with the core of the 
language and for the undemocratic character they give 
to the vocabulary. While the composite character of the 
language gives variety and to some extent precision to the 

1 Life and Letters III. 60. 

2 Born in Exile 380. 

3 Democracy and Libtrty I. p. XXI. 

Summing-up. i^i 

style of the greatest masters, on the other hand it en- 
courages an inflated turgidity of style. Without siding 
completely with Milton's teacher Alexander Gill, who 
says that classical studies have done the English language 
more harm than ever the cruelties of the Danes or the 
devastations of the Normans^, we shall probably be near 
the truth if we recognize in the latest influence from the 
classical languages 'something between a hindrance and 
a help/ 

I Ad Latina venio. Et si uspiam querelas locus, hie est; 
quod otium, quod literae, maiorem cladem sermoni Anglico 
intulerint quam uUa Danorum ssevitia, uUa Normannorum 
vastitas unquam inflixerit. Logonomia Anglica 162 1 (Jiriczek's 
reprint, Strassburg 1903, p. 43.) 

Chapter VII. 
Various Sources. 

151. Although Enghsh has borrowed a great many 
words from other languages than those mentioned in the 
preceding chapters, these borrowings need not occupy 
us long here. For only Scandinavian, French, and Latin 
have left a mark on English deep enough to modify its 
character and to change its structure, and numerous as 
are the words it has borrowed from Dutch, Italian, Span- 
ish, German, etc., the English language would remain 
the same in every essential respect even were they all 
to disappear to-morrow. Many of the words taken over 
from other languages are indeed extremely interesting 
from many points of view, and the student who should 
go through the lists given by Skeat' with a view to 
arranging them in groups according to their signification 
would be able to draw^many important inferences with 
regard to England's commercial and other relations with 
many nations. Attention has already been called to the 
musical terms derived from Italian (§ 31), and a similar 
list of terms of architecture and art in general taken 
from the same language (e. g. colonnade, cornice, corri- 
dor, grotto, niche, parapet, pilaster, profile; miniature, 
fresco; improvisatore, motto) could be made the basis of 

I In his Etymological Dictionary and Principles of English 

Foreign Words. I 53 

an interesting chapter in a history of European civilization, 
A considerable number of military words (e. g. alarm 
or alarum, cartridge, corporal, cuirass, pistol, sentinel) 
carry us back to wars between Italy and France; and 
still other lessons in military history might be learnt from 
the existence in Enghsh of two synonyms, plunder, a 
German word introduced in the middle of the seventeenth 
century by soldiers who had served under Gustavus 
Adolphus, and loot, a Hindi word learnt by English 
soldiers in India about a hundred years ago. But it 
would lead us too far if we were to give many such in- 

152. There is, of course, nothing peculiarly English 
in the adoption of such words as maccaroni and lava 
from Italian, steppe and verst from Russian, caravan and 
dervish from Persian, hussar and shako from Hungarian, 
hey and caftan from Turkish, harem and mufti from 
Arabic, bamboo and orang-outang from Malay, taboo from 
Polynesian, chocolate and tomato from Mexican, moccassin, 
tomahawk, and totem from other American languages. 
As a matter of fact, all these words now belong to the 
whole of the civilized world; like such classical or pseudo- 
classical words as nationality, telegram, and civilization 
they bear witness to the sameness of modern culture 
everywhere : the same products and to a great extent the 
same ideas are now known all over the globe and many 
of them have in many languages identical names. 

153. And yet, English differs from most other languages 
in that it is more inclined than they are to swallow 
foreign words raw, so to speak, instead of preferring to 
translate the foreign expression into some native equi- 
valent. Thus English has taken over the German word 
kindergarten unchanged, while for the same institution 
Danish has the literal translation bornehave and Nor- 
wegian barnehave. 

ICA VII. Various Sources. 

154. An interesting contrast may be seen between the 
behaviour in this respect of the Dutch and the English 
in South Africa. The former, finding there a great many- 
natural objets which were new to them, designated them 
either by means of existing Dutch words whose meanings 
were, accordingly, more or less modified, or else by 
coining new words, generally compounds. Thus shot 
'ditch' was applied to the peculiar dry rivers of that 
country, veld 'field' to the open pasturages, and kopje 
'a little head or cup' to the hills, etc.; different kinds of 
animals were called roodehok ('red-buck'), steenbok ('stone- 
buck'), springbok ('hop-buck'), springhaas ('hop-hare'), 
hartebeest ('hart-beast'); a certain bird was called slang- 
vreter ('serpent-eater'), a certain large shrub spekboom 
('bacon-tree'), etc. The English, on the other hand, 
instead of imitating this principle, have simply taken 
over all these names into their own language, where they 
now figure^ together with some other South African 
Dutch words, among which may be mentioned trek and 
spoor, in the special significations of 'colonial migration' 
and 'track of wild animal', while the Dutch words are I 
much less specialized [trekken 'to draw, pull, travel, 
move'; spoor 'trace, track, rail'). These examples of 
borrowings might easily be multiplied from other do- 
mains, and we may say of the English what Moth says 
of Holofernes and Sir Nathaniel that 'they have been at 
a great feast of languages, and stolne the scraps' (Love's 
L. L. V. I 39). It will therefore be natural to inquire into 
the cause of this linguistic omnivorousness. 

155. It would, of course, be irrational to ascribe the 
phenomenon to a greater natural gift for learning lan- 

I Roodebok often spelt in accordance with the actual Dutch 
pronunciation rooibok, rooyebok. Shot often appears in the un- 
Dutch spelling sluit. 

South Africa. 155 

guages, for in the first place, the English are not usually 
credited with such a gift, and secondly the best linguists 
are generally inclined to keep their own language pure 
rather than adulterate it with scraps of other languages. 
Consequently, we should be nearer the truth if we were 
to give as a reason the linguistic incapacity of the average 
Englishman. As a traveller and a colonizer, however, 
he is thrown into contact with people of a great many 
different nations and thus cannot help seeing numerous 
things and institutions unknown in England. R.L.Steven- 
son says somewhere about the typical John Bull, that 
'his is a domineering nature, steady in fight, imperious 
to command, but neither curious nor quick about the 
life of others'.^ And perhaps the loan-words we are con- 
sidering, testify to nothing but the most superficial curios- 
ity about the life of other nations and would not have 
been adopted if John Bull had really in his heart cared 
any more than this for the foreigners he meets. He is 
content to pick up a few scattered fragments of their 
speech — just enough to impart a certain local colouring 
to his narratives and political discussions, but he goes no 

156. A rather different attitude towards foreign words 
seems to have been taken in former times. On the one 
hand, some foreign place-names of obvious etymology 
were translated; the Black Forest is one of these trans- 
lations which has been retained, while now the Siehen- 
gehirge and the Riesengehirge are terms more commonly 
used than the Seven Mountains and the Giant Mountains. 
On the other hand, the title signior was in the times of 
Shakespeare used very frequently in speaking about 
others than Italians, while now such titles are only applied 
to natives of the country the titles are borrowed from. 

I Memories and Portraits, p. 3. 

ic6 VII. Various Sources. 

It is, indeed, a characteristic feature that foreigners are 
mentioned in England as Signor Manfredini, Herr 
Schultze, Fraulein Adler, etc., who in France would be 
simply Monsieur or Mademoiselle So-and-so. This may 
be interpreted as a sign of a great respect for or deference 
to foreigners, and perhaps that is true in the case of 
foreign musicians or teachers of languages, but in other 
cases, the use of foreign titles may be an outcome of a 
certain unwillingness to recognize foreigners as entitled 
to the same standing as natives, and a consequent in- 
clination to mark them off as un-English. 

157. The tendency to adopt words from other languages 
is due, then, probably to a variety of causes. Foremost 
among these I think it is right to place the linguistic 
laziness mentioned in § 130 and fostered especially by 
the preference for words from the classical languages. 
That the borrowing is not occasioned by an inherent 
deficiency in the language itself, is shown by the ease 
with which new terms actually are framed whenever the 
need of them is really felt, especially by uneducated people 
who are not tempted to go outside their own language 
to express their thoughts. Interesting examples of this 
natural inventiveness may be found in Mr. Edward 
E. Morris's 'Austral English, A dictionary of Australasian 
words, phrases and usages'. As Mr. Morris says in his 
preface, 'Those who, speaking the tongue of Shakespeare, 
of Milton, and of Dr. Johnson, came to various parts of 
Australasia, found a Flora and a Fauna waiting to be 
named in English. New birds, beasts and fishes, new 
trees, bushes and flowers, had to receive names for 
general use. It is probably not too much to say that 
there never was an instance in history when so many 
new names were needed, and that there never will be 
such an occasion again, for never did settlers come, nor 
can they ever again come, upon Flora and Fauna so com- 

Australia. 1 57 

pletely different from anything seen by them before'. 
The gaps were filled partly by adopting words from the 
aboriginal languages, e. g. kangaroo, wombat, partly by 
applying English words to objects bearing a real or fan- 
cied resemblance to the objects denoted by them in Eng- 
land, e. g. magpie, oak, beech, but partly also by new Eng- 
lish formations. Accordingly, in turning over the leaves 
of Mr. Morris's Dictionary we come across numerous 
names of birds like friar-bird, frogsmouth, honey- eater, 
ground-lark, forty-spot^, of fishes like long-fin, trumpeter, 
of plants like sugar-grass, hedge-laurel, ironheart, thousand- 
jacket. Most of these show that the settler must have 
had an imagination. Whip-bird, or Coach-whip, from 
the sound of the note, Lyre-bird from the appearance 
of the outspread tail, are admirable names.' (Morris, /. c.) 
It certainly seems a pity that book-learned people when 
wanting to enrich their mother tongue have not, as a rule, 
drawn from the same source or shown the same talent for 
picturesque and 'telling' designations. 

158. A great many words are nowadays coined by 
tradespeople to designate new articles of merchandise. 
Very little regard is generally paid to correctness of 
formation, the only essential being a name that is good 
for advertizing purposes. Sometimes a mere arbitrary 
collection of sounds or letters is chosen, as in the case of 
kodak, and sometimes the inventor contents himself with 
some vague resemblance to some other word, which may 

I One story of a curious change of meaning must be re- 
counted in Mr. Morris's words : ' The settler heard a bird laugh 
in what he thought an extremely ridiculous manner , its opening 
notes suggesting a donkey's bray — he called it the 'laughing 
jackass'. His descendants have dropped the adjective, and it 
has come to pass that the word 'jackass' denotes to an Australian 
something quite different from its meaning to other speakers 
of our English tongue'. 

J eg VII. Various Sources, 

assist the buyer to remember the name. In one single 
number of one of the illustrated magazines I find the 
following trade names. I add the probable source of any 
name for which I have been able to imagine one: — 
Larola, luxette [luxe], koko, Diano [makes women beauti- 
ful: Diana], melodeon [a musical instrument: melody], 
bath-eucryl [soap, one of the ingredients is ^M^-alyptus], 
oktis, trilene [tablets to cure fat people, try.? or Latin tri 
as in tricolour.? + lean], vapo-cresolene [cresolene va- 
porized], harlene [hair], stenotyper [sort of typewriter for 
stenography], antexema [anti + eczema], mene, vive [a 
photographic camera, cf. vivid], kals [underclothing, cf. 
calegon], nonalton [a tonic, which may be indicated by 
the ending], onomosto, haydal, wincarnis [a tonic: wine, 
caro.?], vinolia [vinum, oleum], bovril [bos, vril, an electric 
fluid in Lytton's novel The Coming RaceY. As the list 
dates from January 1900, a great many of the names will 
probably be extinct before my book sees the light. 
Others may live and even pass into common use outside 
the sphere for which they were originally invented; this 
is the case with kodak. 

159. It once occurred to Mr. Leon Mead to ask a great 
number of the best known American authors and men 
of science what words, if any, they had ever coined. The 
answers he received are very curious^. A great many of 
his correspondents distinctly repudiated the idea of having 
ever done such a thing as coining a word, some explicitly 
declaring that they looked upon the coining of words 
as a crime to be classed with the coining of false money, 
others saying simply that they had always found the 

1 Sometimes these trade names are half-disguised by fancy 
spellings, the Phiteesi boot, Stickphast, Unceda cigar [= you 
need a cigar] in England, Uneeda biscuit in America. 

2 Leon Mead, IVord-Coiftage. New York, Thomas Y. Cro- 
well & Co. 1902. 

Coined Words. I^g 

language of Shakespeare — or some other great author 
they chose to mention — sufficient to express all their 
thoughts. On the other hand, some persons seemed to 
be proud of their coinages and sent Mr. Mead lists of them 
or regretted not being able to remember them. When 
we examine these coined words, we find that by far the 
greater number of them are framed on classical lines, for 
instance lyronym, metropoliarchy, cynophiles, feminology, 
societology, monopolian, hippopcean, to hermetize oneself, 
and deanthropomorphization; I leave out a great many that 
seem still more ugly and unnecessary. Only rarely do we 
come across some word formed by a specifically English 
process, such as densen ('As the spring comes on and the 
densening outHnes of the elm give daily a new design for 
a Grecian urn', Th. W. Higginson), viewpoint and watch- 
point (Fawcett), which are, however, only translations 
from German. Professor Van Dyke says that there was 
once a httle river that could not be described by any 
other adjective than water f ally, and a bird whose song 
seemed to him wild-flowery. The proof-reader objected 
to both of these words, but Dr. Van Dyke withstood him. 
This latter remark is highly characteristic of the attitude 
taken by most professional champions of correctness of 
language towards anything a little out of the common, 
I however justifiable the innovation may be. Very few 
i! people have the courage to say, as Mr. Edgar Fawcett says 
j (p. 82): 'I think every writer ought to have on his con- 
1 science the coining of at least five good [monosyllables] 
j each year.' It may be doubted indeed if the result would 
always be 'good' words, if authors sat down consciously 
to fulfil the duty here prescribed to them, for the secret 
of the thing is that most new words which have come 
to be approved were framed without their originators 
being aware at the moment that they were creating 
anything. There is an interesting passage on p. 80 of the 

j5o ^^^« Various Sources. 

book mentioned: 'He [A. T. Mahan] used once by chance > 
the word eventless — 'dull, weary, eventless month'. 
The word slipped without premeditation off his pen. 
He immediately thought it without authority and found 
it not in Worcester. Nevertheless he stuck to it' as 
briefer, stronger and much more significant than the 
'stupid' uneventful. Now, if people better realized the 
necessary shortcomings and deficiencies of dictionaries, 
they would not go to them as authorities with regard 
to such questions^. A word may have been used 
scores of times without finding its way into any 
dictionary, — and a word may be an excellent one even 
if it has never been used before by any human being. 
If at its first appearance it is just as intelligible as 
if it had been in constant use for centuries, why 
should the first occurrence be more faulty than the 

160. As already hinted, the chief enrichment of the 
language has taken place through those regular processes, 
which are so familiar that any new word formed by| 
means of them seems at once an old acquaintance. Thai 
whole history of English word-formation may be summed 
up thus — that some formative adjuncts have been 
gradually discarded, especially those that presented some 
difficulty of application, while others have been continu- 
ally gaining ground, because they have admitted of being 
added to all or nearly all words without occasioning any 
change in the kernel of the word. Among the former I 
shall mention -en to denote female beings (cf. German -in). 
In Old English this had already become very impractic- 
able because sound changes had occurred which ob- 

I As a matter of fact, Bradley in the N.E.D. quotes Mad. 
DArblay(i8is), Morris (1868), Stanley (1878) and Sherer (1880) 
for eventless, Post (1888) for eventlessly y and Howells (1872) for 


Word -Formation. l6r 

scured the connection between related words. Correspond- 
ing to the masculine j)^gn 'retainer', jj^ow 'slave', wealh 
'foreigner', scealc 'servant', fox, we find the feminine 
Jngnen, ])iewen, wielen, scielcen, fyxen. It seems clear that 
new generations would find some difficulties in forming 
new feminines on such indistinct analogies, so we 
cannot wonder that the ending ceased to be pro- 
ductive. Of the words mentioned, fyxen is the only 
one surviving, and every trace of its connexion with 
fox is now lost, both the form vixen (with its v from 
Southern dialects) and the meaning being now too far 
from the origin. 

161. A much more brilliant destiny was reserved for 
the Old English ending -isc. At first it was added only 
to nouns indicating nations, whose vowel it changed by 
mutation; thus Englisc, now English, from Angle, etc. 
In some adjectives, however, no mutation was possible, 
e. g. Irish, and by analogy the vowel of the primitive 
word was soon introduced into some of the adjectives, 
e. g. Scottish (earlier Scyttisc), Danish (earlier Denisc). 
The ending was extended first to words whose meaning 
was cognate to these national names, heathenish, O.E. 
folcisc or peodisc 'national' (from folc or peod 'people') ; 
then gradually came childish, churlish, etc. Each century 
idded new extensions, foolish and feverish, for instance, 
dating from the fourteenth, and boyish and girlish from 
^he sixteenth century, until now -ish can be added 
:o nearly any noun and adjective (swinish, bookish, 
greenish, biggish, etc.), nay even to whole phrases, 
'^mong recent nonce - formations recorded in the 
M.E.D. may be mentioned 'an I - dont - know - howish- 
less', 'a clean - cravatish formahty of manners', 'Miss 

162. We shall see in a later section (§ 200) that the 
ending -ing has still more noticeably broken the bounds 

Jespbrsen: English. 2ud ed. II 

J 52 ^11- Various Sources. 

of its originally narrow sphere of application. Another 
case in point is the verbal suffix -en. It is now possible to 
form a verb from any adjective fulfilling certain phonetic 
conditions by adding -en (harden, weaken, sweeten, 
sharpen, lessen). But this suffix was not used very much 
before 1500, indeed most of the verbs formed in -en 
belong to the last three centuries. Another extensively 
used ending is -er. Old English had various methods of 
forming nouns to denote agents; from the verb huntan 
'hunt' it had the noun himta 'hunter'; from beodan 'an- 
nounce', boda 'messenger, herald'; from wealdan 'rule', 
wealda; from beran 'bear', bora; from sce])}an 'injure'^ 
scea}a; from weorcan 'work', wyrhta 'wright' (in wheel- 
wright, etc.), though some of these were used in com- 
pounds only; some nouns were formed in -end: rcedenc 
'ruler', scieppend 'creator', and others in -ere: blawen 
'one who blows', blotere 'sacrificer', etc. But it seems aj 
if there were many verbs from which it was impossibh 
to form any agent-noun at all, and the reader will hav( 
noticed that even the formation in a presented somt 
difficulties, as the vowel was modified according to com 
plicated rules. When the want of new nouns was felt 
it was, therefore, more and more the ending -ere that wa: 
resorted to. But the curious thing is that the functioi 
of this ending was at first to make nouns, not from verbs 
but from other nouns, thus O.E. bocere 'scribe' from bo^ 
'book', compare modern hatter, tinner, Londoner, Nei 
Englander, first-nighter. As, however, such a word a; 
fisher, O.E. fiscere, which is derived from the noun a fish 
O.E. fisc, might just as well be analyzed as derived fron 
the corresponding verb to fish, O.E. fiscian, it became 
usual to form new agent-denoting nouns in -er from verbs 
and in some cases these supplanted older formation 
(O.E. hunta, now hunter). Now we do not hesitate t( 
make new words in er from any verb, e. g. a snorer, ; 

Suffixes. 163 

fitter, a telephoner, a total abstainer, etc. Combinations 
with an adverb (a diner-out, a looker-on) go back to 
Chaucer (A somnour is a renner up and down With 
inandements for fornicacioun, D 1284), but do not seem 
to be very frequent before the Ehzabethan period. Note 
also the extensive use of the suffix to denote instruments 
and things, as in slipper, rubber, typewriter, sleeper 
[American = sleeping car). Other much-used suffixes 
cor nouns are : -ness (goodness, truthfulness), -dom 
Christendom, boredom, 'Swelldom', Thackeray), -ship 
ownership, companionship, horsemanship), for adjec- 
:ives: -ly (lordly, cowardly), -y (fiery, churchy, creepy), 
less (powerless, dauntless), - ful (powerful, fanciful), 
ind - ed (blue-eyed , goodnatured , renowned , conceited, 
alented; 'broad - breasted; level - browed , hke the 
Horizon; — thighed and shouldered Hke the billows; 
—footed hke their steahng foam', Ruskin). Prefixes 
i)f wide apphcation are mis-, un-, be-, and others. 
By means of these formatives the Enghsh vocab- 
jlary has been and is being constantly enriched 
jvith thousands and thousands of useful new 

163. There is one manner of forming verbs from nouns 
md vice versa which is specifically English and which is 
)f the greatest value on account of the ease with which 
t is managed, namely that of making them exactly like 
one another. In Old English there were a certain number 
l)f verbs and nouns of the same 'root', but distinguished 
3y the endings. Thus 'I love' through the three persons 
lingular ran lufie lufast lufa}, plural lufiap; the infinitive 
iwas lufian, the subjunctive lufie, pi. lufien, and the im- 
iaerative was lufa, pi. lufia}. The noun 'love' on the other 
laand was lufu, in the other cases lufe, plural lufa or lufe, 
iufum, lufena or lufa. Similarly 'to sleep' was slcepan, 
pres. slcepe slcspest sleep (e)}, slcBpaJi, subjunctive slcEpe, 

II * 


VII. Various Sources. 

slcBpen, imperative sleep, slcepaj), while the noun had the' 
forms sleep, slcepe, and slcepes in the singular, and slcepas, 
slcepum, slcepa in the plural. If we were to give the cor- 
responding forms used in the subsequent centuries, we 
should witness a gradual simplification which had as a 
further consequence the mutual approximation of the 
verbal and nominal forms. The -m is changed into -n, 
all the vowels of the weak syllables are levelled to one 
uniform e, the plural forms of the verbs in -j) give way to 
forms in -n, and all the final n's eventually disappear, 
while in the nouns s is gradually extended so that it be-- 
comes the only genitive and almost the only plural end- 
ing. The second person singular of the verbs retains its 
distinctive -st, but towards the end of the Middle English 
period thou already begins to be less used, and the polite 
ye, you, which becomes more and more universal, claims' 
no distinctive ending in the verb. In the fifteenth cen- 
tury, the e of the endings which had hitherto been pro-; 
nounced, ceased to be sounded, and somewhat later 
became the ordinary ending of the third person singulai ] 
instead of th. These changes brought about the moderr 

scheme: — j 

noun: love loves — sleep sleeps, ! 

verb: love loves — sleep sleeps, \ 

where we have perfect identity of the two parts of speech i 
only with the curious cross-relation between them that .' \ 
is the ending of the plural in the nouns and of the 
singular (third person) in the verbs — an accident 
which might almost be taken as a device for getting 
an s into all indicative sentences containing no pro 
noun (the lover love5; the lovers love) and for showing 
by the place of the 5 which of the two numbers v. 

164. As a great many native nouns and verbs hat 
thus come to be identical in form (e. g. blossom, care 

Nouns and Verbs. 1 65 

deal, drink, ebb, end, fathom, fight, fish, fire), and as the 
same thing happened with numerous originally French 
words (e. g. accord,' O.Fr. acord and acorder, account, 
prm, blame, cause, change, charge, charm, claim, combat, 
comfort, copy, cost, couch), it was quite natural that the 
speech-instinct should take it as a matter of course that 
whenever the need of a verb arose, the corresponding 
noun might be used unchanged, and vice versa. Among 
the innumerable nouns from which verbs have been 
formed in this manner, we may mention a few: ape, awe, 
cook, husband, silence, time, worship. Nearly every word 
for the different parts of the body has given rise to a 
tiomonym verb, though it is true that some of them are 
rarely used: eye, nose (you shall nose him as you go up 
the staires, Hamlet), lip (= kiss, Shakesp.), beard, tongue, 
brain (such stuffe as madmen tongue and braine not; 
Shakesp. Cymbeline), jaw {= scold, etc.), ear (rare, = 
live ear to), chin (American = to chatter), arm {= put 
one's arm round), shoulder (arms), elbow (one's way 
through the crowd), hand, fist (fisting each others throat, 
Shakesp.), finger, thumb, breast {= oppose), body (forth), 
ikin, stomach, limb (they limb themselves, Milton), knee 
'= kneel, Shakesp.), foot. It would be possible in a similar 
ivay to go through a great many other categories of 
KTords; everywhere we should see the same facility of 
"orming new verbs from nouns. 

165. The process is also very often resorted to for 
nonce-words' in speaking and in writing. Thus, a com- 
non form of retort is exemplified by the following quo- 
;ations : Trinkets ! a bauble for Lydia ! ... So this was 
he history of his trinkets! I'll bauble him!' (Sheridan, 
Rivals V. 2). 'I was explaining the Golden Bull to his 
Royal Highness.' T'll Golden Bull you, you rascal!' 
roared the Majesty of Russia (Macaulay, Biographical 
lEss.). 'Such a savage as that, as has just come home 

1 56 VII. Various Sources. 

from South Africa. Diamonds indeed ! I'd diamond hini' 
(Trollope, Old Man's Love) — and in a somewhat different 
manner: 'My gracious Uncle. — Tut, tut, Grace me no 
Grace, nor Uncle me no Uncle' (Shakesp., R 2, cf. also 
Romeo III. 5. 143). 'I heartily wish I could, but — ' 
'Nay, but me no buts — I have set my heart upon it' 
(Scott, Antiq. ch. XI). 'Advance and take thy prize, 
The diamond; but he answered. Diamond me No dia- 
monds ! For God'i love, a little air ! Prize me no prizes, 
for my prize is death' (Tennyson, Lancelot and Elaine). 
166. A still more characteristic peculiarity of the 
English language is the corresponding freedom with which 
a form which was originally a verb is used unchanged 
as a noun. This was not possible till the disappearance 
of the fxnal -e which was found in most verbal forms, 
and accordingly we see an ever increasing number of 
these formations from about 1500. I shall give some 
examples in chronological order, adding the date of the 
earliest quotation for the noun in the N.E.D.: glance 
1503, bend 1529, cut 1530, fetch 1530, hearsay 1532, 
blemish 1535, gaze 1542, reach 1542, drain 1552, gathei 
1555, burn 1563, lend 1575, dislike 1577, frown 1581, 
dissent 1585, fawn (a servile cringe) 1590, dismay 1 590, 
embrace 1592, hatch 1597, dip 1599, dress (persona! 
attire) 1606, flutter 1641, divide 1642, build 1667 (but 
before the nineteenth century apparently used by Pepyj 
only), harass 1667, haul 1670, dive 1700, go 1727 (many ol 
the most frequent applications date from the nineteentl 
century), hobble 1727, lean (the act or condition ol 
leaning) 1776, bid 1788, hang 1797, dig 1819, find 182J 
(in the sense of that which is found, 1847), crave 1830 
kill (the act of killing) 1852, (a killed animal) 1878. I' 
will be seen that the sixteenth century is very fertih 
in these nouns, which is only a natural consequence 0: 
the phonological reason given above. As, however, som( 


Verbs and Nouns. 


of the verb-nouns found in Elizabethan authors have in 
modern times disappeared or become rare, some gram- 
marians have inferred that we have here a phenomenon 
pecuHar to that period and due to the general exuberance 
of the Renaissance which made people more free with 

; their language than they have since been. A glance at 
our list will show that this is a wrong view; indeed, we 

: use a great many formations of this kind which were 
unknown to Shakespeare; he had only the noun a visit- 
ation, where we say a visit, nor did he know our worries, 
our kicks, and moves, etc., etc. 

167. In some cases a noun is formed in this manner 
in spite of there being already another noun derived from 

: the same verb ; thus a move has nearly the same meaning 
as removal, movement or motion (from which latter a new 
verb to motion is formed) ; a resolve and resolution, a laugh 
and laughter are nearly the same thing (though an exhibit 
is only one of the things found at an exhibition). Hence we 

! get a lively competition started between these nouns and 
the nouns in -ing: w^^^ (especially in the sporting world) 
and meeting, shoot and shooting, read (in the afternoon 
I like a rest and a read) and reading^, row (let us go 
out for a row) and rowing (he goes in for rowing), smoke 
and smoking, mend and mending, feel (there was a soft 
feel of autumn in the air. Hall Caine) and feeling. The 
build of a house and the make of a machine are different 
from the building of the house and the making of the 
machine. The sit of a coat may sometimes be spoilt at 
one sitting, and we speak of dressing, not of dress, in 
connexion with a salad, etc. The enormous development 

I Darwin says in one of his letters: 'I have just finished, 
after several reads, your paper'; this implies that he did not 
read it from beginning to end at one sitting ; if he had written 
'after several readings' he would have implied that he had 
read it through several times. . 

1 58 VII. Various Sources. 

of these convenient differentiations belongs to the most 
recent period of the language. Compared with the sets 
of synonyms mentioned above (§ 133: one of the words 
borrowed from Latin, etc.) this class of synonyms shows 
a decided superiority, because here small differences in 
sense are expressed by small differences in sound, and 
because all these words are formed in the most regular 
and easy manner; consequently there is the least possible 
strain put on the memory. 

168. In early English a noun and the verb correspond- 
ing to it] were often similar, although not exactly alike, 
some historical reason causing a difference in either the 
vowel or the final consonant or both. In such pairs of 
words as the following the old relation is kept unchanged : 
a life, to live] a calf, to calve \ a grief, to grieve-, a cloth, 
to clothe', a house, to house; a use, to use — in all these 
the noun has the voiceless and the verb the voiced con- 
sonant. The same alternation has been imitated in a 
few words which had originally the same consonant in 
the noun as in the verb; thus belief, proof, and excuse 
(with voiceless s) have supplanted the older nouns in 
-ve and voiced -se, and inversely the verb grease has now 
voiced 5 [z] where it had formerly a voiceless s. But in 
a far greater number of words the tendency to have 
nouns and verbs of exactly the same sound has prevailed, 
so that we have to knife, to scarf (Shakesp.), to elf 
(id.), to roof, and with voiceless s to loose, to race, to ice, 
to promise, while the nouns repose, cruise (at sea), re- 
prieve, owe their voiced consonants to the corresponding 
verbs. In this way we get some interesting doublets. 
Besides the old noun bath and verb bathe we have the 
recent verb to bath (will you bath baby to-day.?) and 
the noun bathe (I walked into the sea by myself and 
had a very decent bathe, Tennyson). Besides glass 
(noun) and glaze (verb) we have now also glass as a 

Consonants different. 1 69 

verb and glaze as a noun; so also in the case of grass 
and graze, price and prize (where praise verb and 
noun should be mentioned as etymologically the same 

169. The same forces are at work in the smaller class 
of words, in which the distinction between the noun 
and the verb is made by the alternation of ch and k, as 
in speech — speak. Side by side with the old hatch we 
have a new noun a bake, besides the noun stitch and the 
verb stick we have now also a verb to stitch (a book, etc.) 
and the rare noun a stick (the act of sticking); besides 
the old noun stench we have a new one from the verb 
stink. The modern word ache (in toothache, etc.) is a 
curious cross of the old noun, whose spelling has been 
kept, and the old verb, whose pronunciation (with k) 
has prevailed. Baret (1573) says expressly, 'Ake is the 
verb of this substantive ache, ch being turned into k' . 
In the Shakespeare foho of 1623 the noun is always spelt 
with ch and the verb with k; the verb rimes with brake 
and sake. The noun was thus sounded like the name of 
the letter h; and Hart (An Orthographic, 1569, p. 35) 
says expressly, 'We abuse the name of h, calling it ache, 
which sounde serveth very well to expresse a headache, 
or some bone ache.' Indeed, the identity in sound of 
the noun and the name of the letter gave rise to one of 
the stock puns of the time; see for instance Shakespeare 
(Ado III. 4. 56): 'by my troth I am exceeding ill, hey 
ho. — For a hauke, a horse, or a husband.? — For the 
letter that begins them all, H,' and a poem by Heywood: 
'It is worst among letters in the crosse row. For if thou 
finde him other [= either] in thine elbow. In thine arme, 

or leg Where ever you find ache, thou shalt not 

like him.' 

170. Numerous nouns and verbs have the same con- 
sonants, but a difference in the vowels, due either to 


VII. Various Sources. 

gradation or mutation. But here, too, the creative 
powers of language may be observed. Where in old 
times there was only a noun bit and a verb to bite, we 
have now in addition not only a verb to bit (a horse, to 
put the bit into its mouth) as in Carlyle's 'the accursed 
hag 'dyspepsia' had got me bitted and bridled' and in 
Coleridge's witty remark (quoted in the N.E.D.): 'It is 
not women and Frenchmen only that would rather have 
their tongues bitten than bitted', — but also a noun 
bite in various meanings, e. g. in 'his bite is as dangerous 
as the cobra's' (Kipling) and 'she took a bite out of the 
apple' (Ant. Hope). From the noun seat (see above, 
§ 72) we have the new verb to seat (to place on a seat), 
while the verb to sit has given birth to the noun sit 
(cf. § 167). No longer content with the old sale as the 
noun corresponding to sell, in slang we have the new 
noun a (fearful) sell (an imposition); cf. also the Ameri- 
can substantive tell (according to their tell, see Farmer 
and Henley). As knot (n.) was to knit (v.), so was coss 
to kiss, but while of the former pair both forms have 
survived and have given rise to a new verb to knot and 
a new noun a knit (he has a permanent knit of the brow, 
N.E.D.), from the latter the ^-form has disappeared, the 
noun being now formed from the verb: a kiss. We have 
the old brood (n.) and breed (v.), and the new brood (v.) 
and breed (n.); a new verb to blood exists by the side 
of the old to bleed, and a new noun feed by the side of 
the old food. It is obvious that the language has been 
enriched by acquiring all these newly formed words; 
but it should also be admitted that there has been a 
positive gain in ease and simplicity in all those cases 
where there was no occasion for turning the existing 
phonetic difference to account by creating new verbs or 
nouns in new significations, and where, accordingly, one 
of the phonetic forms has simply disappeared, as when 

Vowels different 17 I 

the old verbs sniwan, scry dan, swierman have given way 
to the new snow, shroud, swarm, which are like the 
nouns, or when the noun swat, swot (he swette blodes 
swot, Ancrene Riwle) has been discarded in favour 
of sweat, which has the san^e vowel as the verb. So 
far from the older school of philologists being right 
when they maintained that the formal distinction be- 
tween verbs and nouns was characteristic of the highest 
stage of linguistic development,^ we see that the 
steadily continued approximation of the two classes of 
words has been in English a great aid to linguistic 

171. Among the other points of interest presented by 
the formations occupying us here^ I may mention the 
curious oscillation found in some instances between noun 
and verb. Smoke is first a noun (the smoke from the 
chimney), then a verb (the chimney smokes, he smokes 
a pipe); then a new noun is formed from the verb in the 
last sense (let us have a smoke). Similarly gossip (a) noun: 
godfather, intimate friend, idle talker, (b) verb: to talk 
idly, (c) new noun: idle talk; dart (a) a weapon, (b) to 
throw (a dart), to move rapidly (Hke a dart), (c) a sudden 
motion; brush (a) an instrument, (b) to use that instru- 
ment, (c) the action of using it: your hat wants a brush; 
sail (a) a piece of canvas, (b) to sail, (c) a sailing excursion ; 
wire (a) a metallic thread, (b) to telegraph, (c) a tele- 
gram; so also cable; in vulgar language a verb is formed 
to jaw and from that a second noun a jaw ('what speech 
do you mean?' 'Why that grand jaw that you sputtered 
forth just now about reputation,' F. C. Philips). Some- 
times the starting point is a verb, e. g. frame (a) to 

1 See especially Aug. Schleicher, Die unterscheidung von 
nomen und vefbum, 1865. 

2 On the accent in conduct, to conduct; an object, to object, 
etc. see my Mod. Engl. Grammar, ch. V. 

jy2 ^I'- Various Sources, 

form, (b) noun: a fabric, a border for a picture, etc., 
(c) verb: to set in a frame; and sometimes an ad- 
jective, e. g, faint (a) weak, (b) to become weak, (c) a 
fainting fit. 

172. To those who might see in the obhteration of 
the old distinctive marks of the different parts of speech 
a danger of ambiguity, I would answer that this danger 
is more imaginary than real. I open at random a modern 
novel (The Christian, by Hall Caine) and count on one 
page (173) 34 nouns which can be used as infinitives 
without any change, and 38 verbs the infinitives of which 
are used unchanged as nouns\ while only 22 nouns and 
9 verbs cannot be thus used. As some of the ambiguous 
nouns and verbs occur more than once, and as the same 
page contains adverbs, prepositions, and conjunctions^ 
which can be used as nouns (adjectives) or verbs, or 
both, the theoretical possibiHties of mistakes arising from 
confusion of parts of speech would seem to be very 
numerous. And yet no one reading that page would 
feel the slightest hesitation about understanding every 
word correctly, as either the ending or the context shows 
at once whether a verb is meant or not. Even such an 
extreme case as this line, which is actually found in a 
modern song, 'Her eyes like angels watch them still' is 
not obscure, although her might be both accusative and 
possessive, eyes both noun and verb, like adjective, con- 

1 Answer, brother, reply, father, room, key, haste, gate, 
time, head, pavement, man, waste, truth, thunder, clap, storey, 
bed , book , night , face , point , shame , while , eye , top , hook, 
finger, bell, land, lamp, taper, shelf, church, — whisper, wait, 
return, go, keep, call, look, leave, reproach, do, pass, come, 
cry, open, sing, fall, hurry, reach, snatch, He, regard, creep, 
lend, say, try, steal, hold, swell, wonder, interest, see, choke, 
shake, place, escape, ring, take, light, (I have not counted 
auxiliary verbs.) 

2 Back, down, still, out, home, except, like, while, straight. 

Parts of Speech. iy3 

junction, and verb, watch noun and verb, and still adjec- q-, _ 
tive and adverb. A modern Englishman, realizing the 
great advantage his language possesses in its power of 
making words serve in new functions, might make Shake- 
speare's lines his own in a different sense: 

*So all my best is dressing old words new, 
Spending againe what is already spent ^' 

173. Having thus considered the modes of forming 
new words by adding something to existing words and 
by adding to them nothing at all, we shall end this 
chapter by some remarks on the formation of new words 
by subtracting something from old ones.^ Such 'back- 
formations', as they are very conveniently termed by Dr. 
Murray, owe their origin to one part of a word being 
mistaken for some derivative suffix (or, more rarely, 
prefix). The adverbs sideling, groveling and darkling 
were originally formed by means of the adverbial ending 
-ling, but in such phrases as he walks sideling, he lies 
groveling, etc., they looked exactly like participles in 
-ing, and the consequence was that the new verbs to 
sidle, to grovel, and to darkle were derived from them 
by the subtraction of -ing. The Banting cure was named 
after one Mr. Banting; the occasional verb to hant is, 
accordingly, a back-formation. The ending -y is often 
subtracted; from greedy is thus formed the noun greed 
(about 1600), from lazy and cosy the two verbs laze and 
cose (Kingsley), and from jeopardy (French jeu parti) the 
verb jeopard. The old adjective corresponding to diffi- 
culty was difficile as in French, but about 1600 the adjec- 
tive difficult (= the noun minus y) makes its appearance. 

1 Sonnet 76. 

2 Otto Jespersen, Om subtraktionsdannelser, saerligt pa dansk 
eg engelsk, in Festskrift til Vilh. Thomsen. Copenhagen 1894. 
On the subtraction oi s , as if it were a plural sign, see below, 
5 188. 

jjA VII. Various Sources. 

Puppy from French poupee was thought to be formed by 
means of the petting suffix y, and thus pup was created; 
similarly cad may be from caddy, caddie = Fr. cadet (a 
youngster) and pet from petty = Fr. petit, the transition 
in meaning from 'little' to 'favourite' being easily account- 
ed for. Several verbs originate from nouns in -er (-ar, 
-or)^ which were not originally 'agent nouns'; butcher is 
the French boucher, derived from bouc 'a buck, goat' with 
no corresponding verb, but in English it has given rise 
to the rare verb to butch and to the noun a butch-knife. 
Similarly harbinger, rover, pedlar, burglar, hawker, and 
probably beggar, call into existence the verbs to harbinge 
(Whitman), rove, peddle, burgle, hawk, and beg\ and the 
Latin words editor, donator, vivisector, produce the un-Latin 
verbs to edit, donate (American), vivisect (Meredith), etc. 
which look as if they came from Latin participles.^ 
Some of these back-formations have been more success- 
ful than others in being generally recognized in Standard 

174. It is not usual in Germanic languages to form 
compounds with a verb as the second, and an object, 
an adverb, etc. as the first, part. Hence, when we find 
such verbs as to housekeep (Mrs. Humphrey Ward, Kip- 
ling, Merriman), the explanation must be that -er has 
been subtracted from the perfectly legitimate noun a 
housekeeper (or -ing from housekeeping). The oldest 
examples I know of this formation are to backbite, to 
partake (parttake) and to conycatch (Shakesp.); others 
are to hutkeep, common in Australia, book-keep (Shaw), 
to soothsay, to thoughtread (Why don't they thoughtread 
each other? H. G. Wells), to typewrite (I could typewrite 
if I had a machine, id., also in B. Shaw's Candida), to 
merrymake (you merrymake together, Du Maurier). It 

I Cf. however, my paper quoted above, p. 173. 

Back -Formations. 17 c 

will be seen that most of these are nonce-words. The 
verbs to henpeck and to sunburn are back-formations 
from the participles henpecked and sunburnt] and Brown- 
ing even says 'moonstrike him !' (Pippa Passes) for 'let 
him be moonstruck.' 

175. We have seen (§ 7 ff.) that monosyllabism is one 
of the most characteristic features of modern English, 
and this chapter has shown us some of the morphological 
processes by which the original stock of monosyllables 
has been in course of time considerably increased. It 
may not, therefore, be out of place here briefly to give 
an account of some of the other modes by which such 
short words have been developed. Some are simply 
longer words which have been shortened by regular 
phonetic development (cf. love § 163); e. g. eight 0. E. 
eahta, dear 0. E. deore, fowl O. E. fugol, hawk 0. E. hafoc^ 
lord O. E. hlaford, not and nought O. E. nawiht, pence O.E. 
penigas, ant O. E. cemette, etc. Miss before the names of 
unmarried ladies is a somewhat irregular shortening of 
'missis' (mistress); though found here and there in the 
seventeenth century. Miss was not yet recognized in the 
middle of the eighteenth century (cf. Fielding's Mrs. 
Bridgit, Mrs. Honour, etc.). 

176. This leads us to the numerous popular clippings 
of long foreign words, of which rarely the middle (as in 
Tench 'the House of Detention' and teck 'detective') or the 
end (as in bus 'omnibus', baccer, baccy 'tobacco', phone 
'telephone'), but more often the beginning only subsists. 
Some of the short forms have never passed beyond slang, 
such as sov 'sovereign', pub 'public-house', confab 'con- 
fabulation', pop 'popular concert', vet 'veterinary sur- 
geon', Jap 'Japanese', guv 'Governor', Mods 'Moderations', 
an Oxford examination, matric 'matriculation', prep 'pre- 
paration' and impot or impo 'imposition' in schoolboy's 
slang, sup 'supernumerary', props 'properties' in theatri- 


VII. Various Sources. 

cal s\,^ perks 'perquisites', comp 'compositor', caps 
'capital letters', etc., etc. Some are perhaps now in a fair 
way to become recognized in ordinary speech, such as 
exam 'examination', and bike 'bicycle'; and some words 
have become so firmly established as to make the 
full words pass completely into oblivion, e. g. cab 
(cabriolet), fad (fadaise) , navvy (navigator) and mob 
(mobile vulgus). 

177. A last group of English monosyllables comprises 
a certain number of words the etymology of which has 
hitherto baffled all the endeavours of philologists. At a 
certain moment such a word suddenly comes into the 
language, nobody knowing from where, so that we must 
feel really inclined to think of a creation ex nihilo. I 
am not particularly thinking of words denoting sounds 
or movements in a more or less onomatopoetic way, for 
their origin is psychologically easy to account for, but 
of such words as the following, some of which belong 
now to the most indispensable speech material: bad}-, 
big^, lad and lass, all appearing towards the end of the 
thirteenth century; /i^ adjective and /i/ substantive, prob- 
ably two mutually independent words, the adjective 
dating from 1440, the substantive in the now current 
sense from 1547; dad 'father', jump, crease 'fold, wrinkle', 
gloat, and bet from the sixteenth century; job, fun (and 
pun?), blight, chum and hump from the seventeenth cen- 
tury; fuss, jam verb and substantive, and hoax from the 
eighteenth, and slum perhaps from the nineteenth cen- 
tury. Anyone who has watched small children carefully 
must have noticed that they sometimes create some such 

1 See Zupitza's attempt at an explanation in the NED., which 
does not account for the origin of bceddel. 

2 The best explanation is Bjorkman's, see Scand. Loan- 
Words p. 157 and 259; but even he does not claim to have 
solved the mystery completely. 

Words of Uncertain Origin. I 77 

word without any apparent reason; sometimes they stick 
to it only for a day or two as the name of some plaything, 
etc., and then forget it; but sometimes a funny sound 
takes lastingly their fancy and may even be adopted by 
their playmates or parents as a real word. Without pre- 
tending that such is the origin of all the words just 
mentioned I yet venture to throw out the suggestion that 
some of them may be due to children's playful inven- 

Jespkrshn: Ehiglish. 2ad ed. 12 

Chapter VIII. 

178. The preceding chapter has already brought us 
near to our present province or rather has crossed its 
boundary, for word-formation is rightly considered one 
of the main divisions of grammar. In the other divisions 
a survey of the historical development shows us the same 
general tendency as word-formation does (§ i6o), the 
tendency, as we might call it, from chaos towards cosmos. 
Where the old language had a great many endings, most 
of them with very vague meanings and applications, 
Modern English has but few, and their sphere of signi- 
fication is more definite. The number of irregularities and 
anomahes, so considerable in Old English, has been great- 
ly reduced so that now the vast majority of w^ords are 
inflected regularly. It has been objected that most of 
the old strong verbs are still strong, and that this means 
irregularity in the formation of the tenses: shake shook 
shaken is just as irregular as Old English scacan scoc 
scacen. But it must be remembered, first, that there 
is a complete disappearance of a great many of those 
details of inflexion, which made every Old English para- 
digm much more complicated than its modern successor, 
such as distinctions of persons and numbers, and nearly 
all differences between the infinitive, the imperative, 
the indicative, and the conjunctive, — secondly that 
the number of distinct vowels has been reduced in many 

Simplification. lyg 

verbs; compare thus beran birep beer bceron boren with 
bear bears bore bore born, feohtan (fieht) feaht juhton 
fohten with fight (fights) fought fought fought, bindan band 
bunden with bind bound bound, berstan bcerst burston 
borsten with burst burst burst burst, — and thirdly that the 
consonant change found in many verbs (ceas curon, snaj) 
snidon, teah tugon) has been abolished altogether except 
in the single case of was were. The greatest change to- 
wards simplicity and regularity is seen in the adjectives, 
where one form now represents the eleven different forms 
used by the contemporaries of Alfred. 

179. It would take up too much space here to ex- 
pound in detail the whole process of grammatical devel- 
opment and simplification. It has taken place not sud- 
denly and from one cause, but gradually and from a 
variety of causes. Even such a seemingly small step as 
that by which the inflexion with nominative ye, accusa- 
tive and dative you has given way to the modern use of 
you in all cases, has been the result of the activity of 
many moving forces.^ Nor must it be imagined that the 
development has in every minute particular made for 
progress; nothing has been gained, for instance, by the 
modern creation of mine and thine as absolute possessive 
pronouns by the side of my and thy.^ Sometimes the 
ways by which new grammatical expressions are won are 
rather round-about, and it is only when we compare the 
entire linguistic structure of some remote period with the 
structure in modern times that we observe that the gain 
in clearness and simplicity has really been enormous. I 
shall select a few points of grammar, which seem to me 
illustrative of the processes of change in general, and (as 
regards some of them) of the progressive tendency I 

1 Progress in Language, chapter VII. 

2 lb. p. 68. 


1 80 VIII. Grammar. 

have mentioned. The first point is the development of 
the 5-ending in nouns (where it is now the usual mark 
of the genitive case and of the plural number) and in 
verbs (where it indicates the third person singular of 
the present tense); as the latter ending has prevailed in 
competition with the th-end'mg, the history of th in the 
formation of ordinal numerals will next be considered. 
Then the wonderful enrichment of the language due to 
the extended use of the zwg-ending will be considered, 
and finally some other points will be treated with the 
greatest briefness possible. 

180. (I. The 5-ending in nouns): In Old English the 
genitive was formed in es in most masculines and neu- 
ters, but beside this a variety of other endings were in 
use with the different stems, in -e, in -re, in -an; some 
words had no separate ending in the genitive, and some 
formed a mutation-genitive [boc 'book', gen. bee). Be- 
sides, the genitive of the plural never ended in -s, but 
in -a or -ra or -na {-ena, -ana). With regard to syntax, 
the genitive case filled a variety of functions, possessive, 
subjective, objective, partitive, definitive, descriptive, etc. 
It was used not only to connect two substantives, but 
also after a great number of verbs and adjectives (re- 
joice at, fear, long for, remember, fill, empty, weary, 
deprive of, etc.) ; it sometimes stood before and some- 
times after the governing word. In short, the rules for 
the formation as well as for the employment of that 
case were complicated to a very high degree. But grad- 
ually a greater regularity and simplicity prevailed in 
accidence as well as in syntax; the 5-genitive was extended 
to more and more nouns and to the plural as well as 
the singular number, and now it is the only genitive 
ending used in the language, though in the plural it is 
in the great majority of cases hidden away behind the 
s used to denote the plural number (kings'^ cf. men's). 

The Genitive. 1 8 1 

The position of the genitive now is always immediately 
before the governing word, and this in connexion with 
the regularity of the formation of the case has been in- 
strumental in bringing about the modern group-genitive, 
where the s is tacked on to the end of a word-group with 
no regard to the logic of the older grammar: the King 
of England's power (formerly 'the kinges power of Eng- 
land'), the bride and bridegroom' s return, etc.^ 

181. As for the use* of the genitive, it has been in 
various ways encroached upon by the combination with 
of. First, its use is now in ordinary prose almost re- 
stricted to personal beings, and even such phrases as 
'society's hard-drilled soldiery' (Meredith), where society 
is personified, are felt as poetical; still more so, of course, 
'thou knowst not golds effect' (Sh.) or 'setting out upon 
life's journey' (Stevenson). But in some set phrases 
the genitive is still established, e. g. out of harm's way; 
he is at his wits' (or wit's) end; so also in the stock quo- 
tation from Hamlet, in my mind's eye, etc. Then to in- 
dicate measure, etc.: at a boat's length from the ship, 
and especially time: an hour's walk, a good night's rest, 
yesterday's post; and this is even extended to such pre- 
positional combinations as to-day's adventures, to-morrow's 

182. Secondly, the genitive (of names of persons) is 
now chiefly used possessively, though this word must 
be taken in a very wide sense, including such cases as 
'Shelley's works,' 'Gainsborough's pictures,' 'Tom's ene- 
my', 'Tom's death,' etc. The subjective genitive, too, is in 
great vigour, for instance in 'the King's arrival,' 'the 
Duke's invitation,' 'the Duke's inviting him,' 'Mrs. Poy- 
ser's repulse of the squire' (G. Eliot). Still there is, in 

I See the detailed historical account of the group -genitive. 
Progress in Language p. 279—318. 

jg2 VIII. Grammar. 

quite recent times, a tendency towards expressing the 
subject by means of the preposition by, just as in the 
passive voice, for instance in 'the accidental discovery by 
Miss Knag of some correspondence' (Dickens) ; 'the appro- 
priation by a settled community of lands on the other 
side of an ocean' (Seeley), 'the massacre of Christians by 
Chinese.' — 'Forster's Life of Dickens' is the same thing 
as 'Dickens's Life, by Forster'. The objective genitive 
was formerly much more common than now, the ambi- 
guity of the genitive being probably the reason of its 
dechne. Still, we find, for instance, 'his expulsion from 
power by the Tories' (Thackeray), 'What was thy pity's 
recompense.?' (Byron). 'England's wrongs' generally 
means the wrongs done to England; thus also 'my cosens 
wrongs' in Shakespeare's R 2 IL 3. 141, but 'your foule 
wrongs' (in the same play. III. i. 15) means the wrongs 
committed by you. In 'my sceptre's awe' (ib. I. i. 118) 
we have an objective, but in 'they free awe pays hom- 
age to us' (Hamlet IV. 3. 63) a subjective genitive. But 
on the whole such obscurity will occur less frequently in 
English than in other languages, where the genitive is 
more freely used. 

183. Now, of has so far prevailed that there are very 
few cases where a genitive cannot be replaced by it, and 
it is even used to supplant a possessive pronoun in such 
stock phrases as 'not for the death of me' (cf. Chaucer's 
'the blood of me,' LGW. 848). Of is required in a great 
many cases, such as 'I come here at the instance of your 
colleague. Dr. H. J. Henry Jekyll' (Stevenson), and it is 
often employed to avoid tacking on the s to too long a 
series of words, as in 'Will Wimble's is the case of many a 
younger brother of a great family' (Addison) or 'the 
wife of a clergyman of the Church of England' (Thacke- 
ray), where most Englishmen will resent the iteration of 
of's less than they do the repeated 5' as in Mrs. Brown- 

(y- Phrases. 1 83 

ing's 'all the hoofs Of King Saul's father's asses'. Even 
long strings of prepositions are tolerated, as in 'on the 
occasion of the coming of age of one of the youngest 
sons of a wealthy member of Parliament', or 'Swift's 
visit to London in 1 707 had for its object the obtaining 
for the Irish Church of the surrender by the Crown of 
the First-Fruits and Twentieths' (Aitken) or 'that sub- 
lime conception of the Holy Father of a spiritual king- 
dom on earth under the sovereignty of the Vicar of 
Jesus Christ himself' (Hall Caine). I suppose that very 
few readers of the original books have found anything 
heavy or cumbersome in these passages, even if they 
may here, where their attention is drawn to the gram- 
matical construction. 

184. Speaking of the genitive, we ought also to men- 
tion the curious use in phrases like *a friend of my bro- 
ther's'. This began in the fourteenth century with such 
instances as 'an officere of the prefectes' (Chaucer G 368), 
where officers is readily supplied (= one of the prefect's 
officers) and 'if that any neighebor of mine (= any of 
my neighbours) Wol nat in chirche to my wyf enclyne' 
(id. B 3091); compare also 'ne no-thing of hise thinges 
is out of my power' (id. I 879). In the course of a few 
centuries, the construction became more and more fre- 
quent, so that it has now long been one of the fixtures 
of the English language. The partitive sense is still con- 
ceivable in such phrases as 'an olde religious unckle of 
mine' (Sh., As III. 3. 362) = one of my uncles, though 
it will be seen that it is impossible to analyze it as being 
equal to 'one of my old religious uncles'. The feeling of 
the partitive origin of the construction must, indeed, 
soon have been lost, and the construction was employed 
chiefly to avoid the juxtaposition of two pronouns, 'this 
hat of mine, that ring of yours' being preferred to 'this 
my hat, that your ring', or of a pronoun and a genitive, 


VIII. Grammar. 

as in 'any ring of Jane's', where 'any Jane's ring' or 
'Jane's any ring' would be impossible; compare also 'I 
make it a rule of mine', 'this is no fault of Frank's', etc. 
In all such cases the construction was found so convenient 
that it is no wonder that it should soon be extended 
analogically where no partitive sense is logically possible, 
as in 'nor shall [we] ever see That face of hers againe' 
(Shakespeare, Lear I. i. 267), 'that flattering tongue 
of yours' (As IV. i. 188), 'Time hath not yet so dried 
this bloud of mine' (Ado IV. i. 195), 'If I had such a 
tyre, this face of mine Were full as lovely as is this of 
hers' (Gent. IV. 4. 190), 'this uneasy heart of ours' 
(Wordsworth), 'that poor old mother of his', etc. When 
we now say, 'he has a house of his own', no one ever 
thinks of this as meaning 'he has one of his own houses', 
so that the meaning of the idiom has changed com- 
pletely — a phenomenon of very frequent occurrence in 
the history of all languages. 

185. In the nominative plural the Old EngHsh de- 
clensions present the same motley spectacle as the geni- 
tive singular. Most masculines have the ending as, but 
some have e (Engle, etc.), some a (suna, etc.) and a great 
many an (guman, etc.); some nouns have no ending at 
all, and most of these change the vowel of the kernel 
(fet, etc.), while a few have the plural exactly like the 
singular (hettend). Feminine words formed their plural 
in a (giefa), in e (bene), in an (tungan) or without any 
ending (sweostor; with mutation bee). Neuters had either 
no ending (word) or else u (hofu) or an (eagan). From 
the oldest period the ending as (later es, s) has been 
continually gaining ground, first among those masculines 
that belonged to other declensional classes, later on also 
in the other genders. The aw-ending, which was common 
to a very great number of substantives from the very be- 
ginning, also showed great powers of expansion and at 



one time seemed as likely as (e)s to become the universal 
plural ending. But finally (e)s carried the day, probably 
because it was the most distinctive ending.^ In the be- 
ginning of the modern period eyen, shoon, and hosen, 
housen, peasen still existed, but they were doomed to 
destruction, and now oxen is the only real plural in n 
surviving, for children as well as the biblical kine and 
brethren are too irregular to count as plurals made by 
the addition of n. The mutation plural has survived in 
some words whose signification causes the plural to occur 
more frequently than, or at least as frequently as, the 
singular: geese^ teeth, feet, mice, lice, men and women. In 
all other words the analogy of the plurals in s was too 
strong for the old form to be preserved. 

186. Instead of the ending -ses we often find a single s\ 
in some cases this may be the continued use of the 
French plural form without any ending [cas sg. and pi.), 
as in sense (their sense are shut, Sh.), corpse (pi. Sh.) etc. 
In Coriolanus III. i. 118 voyce and voyces occur, both of 
them to be read as one syllable: 'Why shall the people 
give One that speakes thus, their voyce .^ — He give my 
reasons. More worthier than their voyces. They know 
the corne.' But when Shakespeare uses princesse, balance, 
or merchandize as plurals (Tp. I. 2. 173; Merch. IV. i. 255; 
Ant. 11. 5. 104), the forms admit of no other explanation \ 1 
\than that of haplology (pronouncing the same sound I I 
bnce instead of twice). Thus also in the genitive case: ^ * 
'his mistresse eye-brow' (As II. 7. 149), 'your High- 
ness' pleasure', etc. Now it is more usual to give the 
full form mistress's, etc., yet in Pears' soap the juxta- 
position of three s'ts is avoided by means of the apostro- 
phized form. The genitive of the plural is now always 
haplologized : 'the Poets' Corner', except in some dialects: 

I Progress in Lmiguage , p. 178 ff. 

I 85 VIII. Grammar. 

'other folks's children' (George Eliot), 'the bairns's clease' 
(Murray, Dial, of Scotl. 164). Wallis (1653) expressly 
states that the gen. pi. in the Lords' House (by him 
written Lord's) stands instead of the Lords' s House (duo \\ 
s in unum coincidunt). A phenomenon of the same order 
is the omission of the genitive sign before a word 
beginning with 5, now chiefly before sake: for fashion 
sake, etc. 

187. Sometimes an s belonging to the stem of the 
word is taken by the popular instinct to be a plural 
ending. Thus in alms (ME. almesse, elmesse, pi. al- 
messes; OE. celmesse from Gr. eleemosune) ; it is signi- 
ficant that the word is very often found in connexions 
where it is impossible from the context to discover 
whether a singular or a plural is intended (ask alms, 
give alms, etc.). In the Authorized Version the word 
occurs eleven times, but eight of these are ambiguous, 
two are clearly singular (asked an almes, gave much 
almes) and one is probably plural (Thy praiers and thine 
almes are come up). Nowadays the association between 
the 5 of the alms and the plural ending has become so 
firm that an alms is said and written very rarely indeed, 
though it is found in Tennyson's Enoch Arden. Riches 
is another case in point; Chaucer still lays the stress on 
the second syllable [richesse as in French) and uses the 
plural richesses; but as subsequently the final e disap- 
peared, and as the word occurred very often in such a 
way that the context does not show its number ('Thou 
bearst thy heavie riches but a journie', Sh. Meas. III. 
I. 27; thus in fourteen out of the 24 places where Shake- 
speare uses it), it is no wonder that the form was gener- 
ally conceived as a plural, thus 'riches are a power' 
(Ruskin). The singular use (the riches of the ship is 
come on shore, Sh. 0th. II. i. 83, too much riches, R2 
III. 4, 60) is now wholly obsolete. i. 

Back -Formations. 1 87 

i 188. A further step is taken in those words that lose 
I the s originally belonging to their stem, because it is 
: mistakenly apprehended as the sign of plural. ^ Latin 
pisum became in OE. pise, in ME. pese, pi. pesen; Butler 
(1633) still gives peas as sg. and peasen as pi., but he 
: adds, 'the singular is most used for the plural: as . . a 
peck of peas; though the Londoners seem to make it a 
regular plural, calling a peas a pea*. In compounds like 
; peaseblossom, peaseporridge and pease- soup (Swift, Ch. 
i Lamb) the oldCfroru; was preserved long after pea had 
I become the recognized singular. Similarly a cherry was 
j evolved from a form in 5 (French cerise), a riddle from 
L: riddles; an eaves (OE. efes, cf. Got. ubizwa, ON. ups) is 
t often made an eave, and vulgarly a pony shay is said for 
\ chaise; compare also Bret Harte's 'heathen Chinee' and 
the parallel forms a Portuguee, a Maltee. An interesting 
case in point is Yankee, according to the highly probable 
explanation recently set forth by H. Logeman. The term 
was originally applied to the inhabitants of the Dutch 
colonies in North America (New Amsterdam, now New 
York, etc.). Now Jan Kees is a nickname still applied 
in Flanders to people from Holland proper. Jan of 
course is the common Dutch name corresponding to 
English John, and Kees may be either the usual pet-form f / 
of the name Cornelis, another Christian name typical of 
the Dutch, or else a dialectal variation of kaas 'cheese' 
in allusion to that typically Dutch product, or — what is 
most probable — a combination of both. Jankees in 
English became Yankees, where the s was taken as the 
plural ending and eventually disappeared, and Yankee 
became the designation of any inhabitant of New Eng- 
land and even sometimes of the whole of the United 

I Cf. the other back-formations mentioned above , S i73- Other 
instances will be found in the paper there quoted. 


jgg VIII. Grammar. 

189. We have a different class of back-formations in 
those cases in which the s that is subtracted is really the 
plural ending, while one part of the word is retained 
which is logically consistent with the plural idea only. 
It is easily conceivable that most people ignorant of the 
fact that the first syllable of cinque-ports means 'five', 
have no hesitation in speaking of Hastings as a cinque- 
port; but it is more difficult to see how the signification 
of the numeral in ninepins should be forgotten, and yet 
sometimes each of the 'pins' used in that play is called 
a ninepin, and Gosse writes 'the author sets up his four 

190. In some words the s of the plural has become 
fixed, as if it belonged to the singular, thus in means. 
As is shown by the pun in Shakespeare's Romeo 'no 
sudden meane of death, though nere so meane' the old 
form was still understood in hi! time, but the modern 
form too is used by him {by that meanes, Merch. ; a means, 
Wint.). Similarly: too much pains, an honourable amends, a 
shambles, an innings, etc., sonietimes a scissors, a tweezers, 
a barracks, a golf links, etc., where the logical idea of a 
single action or thing has proved stronger than the 
original grammar. 

191. It is not, however, till a new plural has been 
formed on such a form that the transformation from 
plural to singular has been completed. This phenomenon^ 
which might be termed plural raised to the second power, 
will naturally occur with greater facility when the original 
singular is not in use or when the manner of forming the 
plural is no longer perspicuous. Thus OE. broc formed 
its plural brec {cf. gos ges goose geese), but broc became 
obsolete, and brec, breech was free to become a singular 
and to form a new plural breeches. Similarly invoices, j, 
quinces, bodices and a few others have a double plural ■' 
ending; but then the unusual sound of the first ending 

Double Plurals. i8q 

(voiceless s, where the ordinary ending is voiced, as in 
joys, sins) facilitated the forgetting of the original function 
of the s (written -ce). Bodice is really nothing but a by- 
form of bodies. The old pronunciation of bellows and 
gallows had also a voiceless s, which helps to explain the 
vulgar plurals bellowses and gallowses. But in the occasional 
plural mewses (from a mews, orig. a mue) the new 
ending has been added in spite of the first 5 being voiced. 
These plurals raised to the second power, to which must 
be added sixpences, threepences, etc., are particularly 
interesting because there really are cases where the want 
is felt of expressing the plural of something which is in 
itself plural, either formally or logically; cf. many (pairs 
of) scissors. Generally one plural ending only is used^, 
but occasionally the logically correct double ending is 
resorted to, especially among uneducated persons,* 
Thackeray makes hii flunkey write: 'there was 8 sets of 
^hamberses" (Yellowplush Papers, p. 39), and a London 
schoolboy^ once wrote: 'cats have clawses' (one cat has 
<:laws !) and again 'cats have 9 liveses' (each cat has nine 
lives !). Dr. Murray^ mentions a double plural sometimes 
formed in Scotch dialect from luch words as schuin (one 
person's shoes), feit 'feet' and kye *cow?5', schuins meaning 
more than one pair of shoes, and he ingeniously suggests 
that this may illustrate such plurals as children, brethren, 
kine; the original plurals were childer, brether, ky (still 
preserved in the northern dialect), which may have 
'come to be used collectively for the offspring or members 
of a single family, the herd of a single owner, so that a 

1 'Then ensued one of the most lively ten minutes that I 
•can remember' (Conan Doyle), plural of 'one ten minutes'. 

2 Very Original English, ed. by Barker (London 1889), 


3 Dialect of the Southetn Counties of Scotland (London 1873), 

3>. 161. 

J go VIII. Grammar. 

second plural inflection became necessary to express the 
brethren and children of many families, the ky-en of many 
owners ... In modern English we restrict brothers, which re- 
places brether, to those of one family, using brethren for those 
who call each other brother, though of different families.' 
192. Most of the words that make their plural like the 
singular are old neuters, the 5-ending belonging originally 
to mascuHnes only and having only gradually been ex- 
tended to the other two genders; thus swine, deer, sheep. 
But as the unchanged plurals were used chiefly in a 
collective sense, a difference sprang up between a collec- 
tive plural (unchanged) and an individual plural (in -5), 
as seen most clearly in Shakespeare's 'Shee hath more 
haire then wit, and more faults then hairs' (Gent. III. 
I. 362) and Milton's 'which thou from Heaven Feigndst 
at thy birth was giv'n thee in thy hair, Where strength 
can least abide, though all thy hairs Were bristles' (Sams. 
Ag. 1 136). This difference was transferred to some old 
masculines, like fish, fowl; and a great m.any names of' 
particular fishes and birds, especially those generally 
hunted and used for food, are now often unchanged in' 
the plural {snipe, plover, trout, salmon, etc.), though ' 
with a great deal of vacillation. It is also noticeable that ' 
much fruit = many fruits and much coal = many coals. 
When we say 'four hundred men', but 'hundreds of men', 
'two dozen collars', but 'dozens of collars' and similarly 
with couple, pair, score and some other words, we have 
an approach to the rule prevailing in many languages, 
e. g. Magyar, where the plural ending is not added after 
a numeral, because that suffices in itself to show that a 
plural is intended. 

193. (II) We proceed to that verbal ending which is 
now identical in form with the ordinary genitival and 

Third Singular. I g I 

plural ending in the nouns, namely s (he loves, etc.). 
In Old-English -th (]?) was used in the ending of the third 
person singular and in all persons in the plural of the 
present indicative, but the vowel before it varied, so that 
we have for instance : — 

Infinitive '^rd sg. pi. 


binde)?, bint 





But in the Northumbrian dialect of the tenth century 
5 was substituted to )? (singular hindes, plural bindas), 
and as all unstressed vowels were soon after levelled, 
the two forms became identical {hindes). As in the same 
dialect the second person singular too ended in 5 (as 
^against the -st of the South), all persons sounded alike 
except the first singular. But the development was not 
to stop there. In Old English a difference is made in the 
plural, according as the verb precedes we or ge ('y^') or 
not {binde we, binde ge, but we binda}, ge bindaji). This 
is the germ of the more radical difference now carried 
through consistently in the Scotch dialect, where the s 
is only added when the verb is not accompanied by its 
proper pronoun, — but in that case it is used in all per- 
sons. Dr. Murray gives the following sentences among 

aa cmn fyrst — yt's mey at cums fyrst. 
wey gang theare — huz tweae quheyles gangs theare. 
they cum an' teake them — the burds cums an' pcecks them. 

I Dial, of the Southern Counties of Scotland, 1873, 
p. 212, where quotations from the earlier literature are also 


VIII. Grammar. 

(I come first; it is I that come first; we go there; we two 
sometimes go there; they come and take them; the birds I 
come and pick them). 

In the other parts of the country the development i 
was different. In the Midland dialect the -en of the sub- 1 
junctive and of the past tense was transferred to the : 
present of the indicative, so that we have the following 
forms in the standard language: — X 

14 th century 16 th cent. 

I falle I fall 

he falleth he fall(e)th 

we fallen (falle) we fall. 

This is the only dialect in which the third person sin- 
gular is kept clearly distinct from the other persons. 

In the South of England, finally, the th was preserved 
in the plural, and was even extended to the first person 
singular. Old people in the hilly parts of Somersetshire 
and Devonshire still say not only [i wo'k)?] 'he walks', 
but also [^ei ze)?, ai ze)?] 'they say, I say'. In most cases, 
however, do is used, which is made [da] without any th 
through the whole singular as well as plural. 

194. But the northern s'ts wandered southward. A 
solitary precursor is found in Chaucer, who writes oncej 
telles instead of the usual telleth for the sake of the rimej 
(:elles, Duchesse 73).^ A century later Caxton used thei 
/^-ending [eth, ith, yth) exclusively, and this remained; 
the usual practice till late in the i6th century, when s 
was first introduced by the poets. In Marlowe s is by 
far the commoner ending, except after hissing consonants 

1 Elworthy, Grammar of the Dialect of West Sonierset,{ 

p. 191 ff. 

2 In the Reves Tale the j- forms are used to characterizei 
the North of England dialect of the two students [gas for Chaucer's 
ordinary gooth, etc.) 

Th and s. 1^3 


I (passeth, opposeth, pitcheth, presageth, etc., Tambur- 
I laine 68, 845, 1415, 1622). Spenser prefers 5 in poetry. 
I In the first four cantos of the Faerie Queene I have counted 
}i94 5'es as against 24 th's (besides 8 has, 18 hath, 15 does, 
and 31 doth). But in his prose th predominates even much 
more than 5 does in his poetry. In the introductory letter 
to Sir W. Raleigh there is only one s (it needs), but many 
th's\ and in his book on 'the Present State of Ireland' all 
the third persons singular end in th, except a small num- 
ber of phrases [me seems, several times, but it seemeth; 
what hoots it; how comes it, and perhaps a few more) that 
seem to be characteristic of a more colloquial tone than 
lithe rest of the book. Shakespeare's practice is not easy 
ito ascertain. In a great many passages the folio of 1623 
fjhas th where the earlier quartos have 5. In the prose 
i parts of his dramas s prevails^ and the rule may be laid 
['down that th belongs more to the solemn or dignified 
i speeches than to everyday talk, although this is by no 
ji means carried through everywhere. In Macbeth I. 7. 29 ff. 
ijLady Macbeth is more matter-of-fact than her husband 
j}(Lady: He has almost supt .. , . Macb.: Hath he ask'd 

[;for me.? Lady: Know you not he ha's. Macb He 

\]hath honour'd me of late ....), but when his more solemn 
■jmood seizes her, she too puts on the buskin (Was the 
[hope drunke, Wherein you drest your selfe.? Hath it 
[slept since.?). — Where Mercutio mocks Romeo's love- 
isickness (II. I. 15), he has the Hne: He heareth not, he 
\stirreth not, he moveth not, but in his famous description 
bf Queen Mab (I. 4. 53 ff.) he has 18 verbs in 5 and only 
•two in th, hath and driveth, of which the latter is used for 

I ' 

ithe sake of the metre. 


I Franz, Shakespeare- Granunatik, 2nd ed. p. 151: In Much 
Ado (Q 1600) th is not found at all in the prose parts and only 
:wice in the poetical parts; the Merry Wives, which is chiefly 

I" in prose , has only one th. 

I QA VIII. Grammar. 

195. Contemporary prose has nearly exclusively th] 
the 5-ending is not at all found in the Authorized Version 
of i6ii, nor in Bacon's Atlantis (though in his Essays] 
there are some s'es). The conclusion with regard to! 
Elizabethan usage as a whole seems to be that the form; 
in ^ was a colloquialism and as such was allowed in poetry I 
and especially in the drama. This s must, however, be 
considered a poetical licence wherever it occurs in that 
period. But in the first half of the seventeenth century 5 
must have been the ending universally used in ordinary 
conversation, and we have evidence that it was even 
usual to read s where the book had th, for Richard Hodges 
(1643) gives in his list of words pronounced alike though 
spelt differently among others boughs boweih howze; clause 
claweth claws; courses courseth corpses; choose cheweth^, 
and in 1649 ^^^ says 'howsoever wee write them thus, 
leadeth it, maketh it, noteth it, we say lead's it, make's 
it, note's it!' The only exceptions seem to have been 
hath and doth, where the frequency of occurrence pro- 
tected the old forms from being modified analogically^ 
so that they were prevalent till about the middle of th 
eighteenth century. Milton, with the exceptions just 
mentioned, always writes 5 in his prose as well as in his 
poetry, and so does Pope. No difference was then felt 
to be necessary between even the most elevated poetry 
and ordinary conversation in that respect. But it is 
well worth noting that Swift, in the Introduction tO; 
his 'Polite Conversation', where he affects a quasi- 
scientific tone, writes hath and doth, while in the 
conversations themselves has and does are the forms 
constantly used.^ 

1 See Ellis, Early English Pronu?iciatiofi , IV, ioi8. 

2 This applies, partially at least, to saith as well. 

3 In the Journal to Stella all verbs have s, except hath, 
which is, however, less common than has. 

Th and s. 195 

196. At church, however, people went on hearing the 
//i-forms, although even there the 5'es began to creep 
in.^ And it must certainly be ascribed to influence from 
bibhcal language that the ^A-forms began again to be 
used by poets towards the end of the eighteenth century; 
at first apparently this was done rather sparingly, but 
nineteenth century poets employ th to a greater^extent. 
This revival of the old form affords the advantage from 
the poet's point of view of adding at discretion a syllable, 
as in Wordsworth's 

In gratitude to God , Who feeds our hearts 

For His own service; knoweth , loveth us (Prelude 13.276) 

or in Byron's 

Whate'er she loveth, so she Imjes thee not, 

What can it profit thee? (Heaven and Earth I sc. 2) 

Sometimes the ih-ioxm comes more handy for the 
rime (as when saith rimes with death), and sometimes the 
following sound may have induced a poet to prefer one 
or the other ending, as in 

Coleridge hath the sway, 

And Wordsworth has supporters, two or three, ^ 

but in a great many cases individual fancy only decides 
which form is chosen. In prose, too, the th-ioxm. begins to 
make its re-appearance in the nineteenth century, not 
only in biblical quotations, etc., but often with the sole 
view of imparting a more solemn tone to the style, as 

in Thackeray's 'Not always doth the writer know whither 



1 See the Spectator, no. 147 (Morley's ed. p. 217) 'a set of 
I readers [of prayers at church] who affect, forsooth, a certain 

gentleman -like familiarity of tone, and mend the language as 
they go on, crying instead of pardoneth and absolveth, pardons 
and absolves.' 

2 Do7i Juan XI, 69. 



VIII. Grammar. 

the divine Muse leadeth him.' Some recent novelists 
affect this archaic trick usque ad nauseam. 

197. The nineteenth century has even gone so far as 
to create a double-form in one verb, making a distinction 
between doth [pronounced dA)?] as an auxiliary verb and 
doeth [pronounced du'i)?] as an independent one. The 
early printers used the two forms indiscriminately, or 
rather preferred doth where doeth would make the line 
appear too closely packed, and doeth where there was 
room enough. Thus in the Authorized Version of 161 1 
we find 'a henne doeth gather her brood under her wings' 
(Luke XIII. 34) and 'he that doth the will of my father' 
(Matth. VII. 21), where recent use would have reverseH 
the order of the forms, but in 'whosoever heareth these 
sayings of mine, and doeth them' (Matth. VII. 24) the 
old printer happens to be in accordance with the rule 
of our own days. When the ^^-form was really livinp^, 
doeth was certainly always pronounced in one syllable 
(thus in Shakespeare). I give a few exam.ples of the modern 
differentiation.^ J. R. Lowell writes (My Love, Poems 
1849, I 129 = Poetical Works in one volume p. 6) 'She 
doeth little kindnesses . . . Her life doth rightly harmonize 
. . . And yet doth ever flow aright.' Rider Haggard has 
both forms in the same sentence (She 199) 'Man doeth 
this and doeth that, but he knows not to what ends his 
sense doth prompt him'; cf. also Tennyson's The Captain: 
'He that only rules by terror, Doeth grievous wrong.' 

198. To sum up. If the s of the third person singular 
comes from the North, this is true of the outer form 
only; the 'inner form', to use the expression of some 
German philologists, is the Midland one, that is to say. 

I Which has not been noticed in Murray's Dictionary, though 
he mentions the corresponding difference between dost and doest 
as 'in late use'. 

Doeth. Numerals. igy 

s is used in those cases only where the Midland dialects 
had th, and is not extended according to the northern 
rules. In vulgar English of the last two centuries s has 
been used in the first person singular: / wishes; says /, 
etc. The oldest instance I have noted is from the Rehear- 
sal (1671) : 'I makes 'em both speak fresh' (Arber's reprint, 
p. 53). But it will be seen that this is in direct opposition 
to the northern usage where the s is never found by the 
side of the personal pronoun.^ 

199. (HI. The ending th in ordinals). While the cardinal 
numerals show very little change during the whole life 
of the language except what is a consequence of ordinary 
phonetic development^, the ordinals have been much 
more changed so that their formation is now completely 
regular, with the exception of the first three. First has 
ousted the old forma (corresponding to Latin primus), 
.nd the French second has been called in to relieve other 
of one of its significations, so that a useful distinction has 
been created between the definite and the indefinite 
numeral. As for the numbers from 4 upwards, the 
regularization has affected both the stem and the ending 
of the numeral. In Old English the n had disappeared 
from seofo^a, nigo^a and teotia (feowerteo^a, etc.), but now 
it has been analogically reintroduced: seventh, ninth, 
tenth (fourteenth, etc.), the only survival of the older forms 
being tithe, which is now a substantive differentiated 
from the numeral, as seen particularly clearly in the phrase 

1 I leave out of consideration the occasional Shakespearian 
s in the plural of the verb as too dubious to be treated in a 
work of this character. 

2 Note that in Old and Middle English the cardinals had 
an -e when used absolutely (y^men; they were Jive], and that 
It is this form that has prevailed. If the old conjoint form had 
survived, ^ve, and Hvelve would have ended in/ and seven, 
nine, ten and eleven would have had no -n. 

iq8 VIII. Grammar. 

*a tenth part of the tithe' (Auth. Version, Num. i8. 26). 
In twelfth and fifth we have the insignificant anomaly of / . 
(which in the former is often mute) instead of v, and the 
consonant-group in the latter has shortened the vowel, 
but elsewhere there is complete correspondence between 
each cardinal and its ordinal. As for the ending, it used 
according to a well-known phonetic rule to be -ta (later 
'te, t) after voiceless open consonants, thus fifta fift, sixta 
sixt, twelfta twelft; and these are still the only forms in 
Shakespeare (Henry the Fift, etc.)^ and Milton. The 
regular forms in th evidently were used in writing before ] 
they became prevalent in speaking, for Schade in 1765 j 
laid down the rule that th was to be pronounced t in 
twelfth and fifth. Eighth, which would be more ade- : 
quately written eightth, is also a modern form; the old 
editions of Shakespeare have eight. The formation in -th, 
which is now beautifully regular, has also been extended 
in recent times to a few substantives: the hundredth, 
thousandth, millionth, and dozenth. 

200. (IV) The history of the forms in ing is certainly 
one of the most interesting examples of the growth from 
a very small beginning of something very important in 
the economy of the language. The 'ing', as I shall for 
shortness call the form with that ending, began as a pure 
noun^, restricted as to the number of words from which 
it might be formed and restricted as to its syntactical 
functions. It seems to have been originally possible tc 
form it only from nouns, cf. modern words like schooling , 
shirting, stabling; as some of the nouns from which ings 

1 Twelfth Night is in the folio of 1623 called Tivelfe NighA 
and similarly we have twelfe day , where the middle consonantl 
of a difficult group has been discarded, just as in the thousand! 
part (As IV. i. 46). A 

2 The Old English ending was ting as well as ing. 

Numerals. Ing. ign 

were derived, had corresponding weak verbs, the ings 
came to be looked upon as derived from these verbs, 
i and new ings were made from other weak verbs, (Also 
! from French verbs, cf. above § io6). But it was a long 
time before ings were made from strong verbs; a few 
occur in the very last decades of the Old English period, 
but most of them did not creep into existence till the 
twelfth or thirteenth century or even later, and it is not, 
perhaps, till the beginning of the fifteenth century that 
the formation had taken such a firm root in the language 
that an ing could be formed unhesitatingly from any 
verb whatever (apart from the auxiliaries can^ may, shall, 
need, etc., which have no ings). 

201. With regard to its syntactical use the old ing was 
a noun and was restricted to the functions it shared with 
all other nouns. While keeping all its substantival qual- 
ities, it has since gradually acquired most of the functions 
belonging to a verb. It was, and is, inflected like a noun; 
now the genitive case is rare and scarcely occurs outside 
of such phrases as 'reading for reading's sake'; but the 
plural is common: his comings and goings; feelings, 
drawings, leavings, weddings, etc. Like any other noun 
it can have the definite or indefinite article and an ad- 
jective before it: a beginning, the beginning, a good be- 
ginning, etc., so also a genitive: Tom's savings. It can 
enter into a compound noun either as the first or as the 
second part: a walking-stick; sight-seeing. The ing can 
be used in a sentence in every position occupied by an 
ordinary noun. It is the subject and the predicative 
nominative in 'complimenting is lying', the object in 'I 
hate lying'; it is governed by an adjective in 'worth 
knowing', and governed by a preposition in 'before an- 
swering', etc. But we shall now see how several of the 
peculiar functions of verbs are extended to the ing. The 
coalescence in form of the verbal noun and of the present 

200 VIII, (jrammar. 

participle is, of course, one of the chief factors of this 

202. When the ing was a pure noun the object of the 
action it indicated could be expressed in one of three 
ways: it might be put in the genitive case ('sio feding 
)7ara sceapa', the feeding of the sheep, Alfred), or it might 
form the first part of a compound (blood-letting) or — 
the usual construction in Middle English — it might be 
added after of (in magnifying of his name, Chaucer). 
The first of these constructions has died out; the last is 
in our days especially frequent after the article (since 
the telling of those little fibs, Thackeray). But from the 
fourteenth century we find a growing tendency to treat 
the ing like a form of the verb and, accordingly, to put 
the object in the accusative case. Chaucer's words 'in 
getinge of your richesses and in usinge hem' (B 2813) 
show both constructions in juxtaposition; so also 'Thou 
art so fat-witted with drinking of olde sacke, and un- 
buttoning thee after supper' (Henry IV, A. I. 2. 2.) 
Chaucer's 'In lif tinge up his hevy dronken cors' (H 67) 
shows a double deviation from the old substantival con- 
struction, for an ordinary noun cannot in this way be 
followed by an adverb, and in the old language the 
adverb was joined to the ing in a different way (up- 
lifting, in-coming, down -going). In course of time it 
became more and more usual to join any kind of ad- 
verb to the ing, e. g. 'a man shal not wyth ones [once] 
over redyng fynde the ryght understandyng' (Caxton), 
'he proposed our immediately drinking a bottle to- 
gether' (Fielding), 'nothing distinguishes great men 
from inferior men more than their always, whether in 
Hfe or in art , knowing the ways things are going' (Ruskin). 

203. A noun does not admit of any indication of time; 
his movement may correspond in meaning to 'he moves 
(is moving)', 'he moved (was moving)', or 'he will move.' 

Ing. 201 

Similarly the ing had originally, and to a great extent 
still has, no reference to time: 'on account of his coming' 
may be equal to 'because he comes' or 'because he came' 
or 'he will come', according to the connexion in which it 
occurs. 'I intend seeing the king' refers to the future, 
'I remember seeing the king' to the past, or rather the 
ing as such implies neither of these tenses. But since the 
end of the sixteenth century the ing has still further 
approximated to the character of a verb by developing 
a composite perfect. Shakespeare, who uses the new 
tense in a few places, e. g. Gent. I. 3. 16 (To let him spend 
his time no more at home; Which would be great im- 
peachment to his age, In having knowne no travaile in 
his youth') does not always use it where it would be used 
now; for in 'Give orders to my servants that they take 
No note at all of our being absent hence' being corresponds 
in meaning to having been, as shown by the context 
(Merch. of Ven. V. 120). — Like other nouns the ing was 
also at first incapable of expressing the verbal distinction 
between the active and the passive voice. The simple ing 
is still often neutral in this respect, and in some con- 
nexions assumes a passive meaning, as in 'it wants 
mending', 'the story lost much in the teUing'. This is 
extremely frequent in old authors, e. g. 'Use everie man 
after his desart, and who should scape whipping' (Ham- 
let II. 2. 554). 'Shall we . . . excuse his throwing into the 
water?' (Wiv. III. 3. 206 = his being, or having been, 
thrown), 'An instrument of this your calling backe' (0th. 
IV. 2. 45). But about 1600 a new form came into existence, 
as the old one would often appear ambiguous, and it 
was felt convenient to be able to distinguish between 
'foxes enjoy hunting' and 'foxes enjoy being hunted'. 
The new passive is rare in Shakespeare ('I spoke ... of 
being taken by the insolent foe', 0th. I. 3. 136), but has 
now for a long time been firmly established in the language. 

20 2 VIII. Grammar. 

204. The last step in this long development of a form 
at first purely substantival into one partly substantival 
and partly verbal in function was taken about two hun- 
dred years ago. The subject of the ing, like that of any 
verbal noun (for instance Ccesar's conquests, Pope's 
imitations of Horace), is for the most part put in the 
genitive case — nearly always when it is a personal pro- 
noun (in spite of his saying so), and generally when it 
indicates a person (in spite of John's saying so). But a 
variety of circumstances led to the adoption in many 
instances of a new construction, which is wrongly taken 
by most grammarians as containing the present participle 
and not the 'gerund'. I shall give elsewhere my reasons 
for not accepting that view and here content myself with 
quoting a few instances of the new construction out of 
several hundreds which I have collected: 'When we talk 
of this man or that woman being no longer the same 
person' (Thackeray), 'besides the fact of those three 
being there, the drawbridge is kept up' (Anth. Hope), 
'When I think of this being the last time of seeing you' 
(Miss Austen), 'the possibility of such an effect being 
wrought by such a cause' (Dickens), 'he insisted upon 
the Chamber carrying out his policy' (Lecky), 'I have not 
the least objection in life to a rogue being hung' (Thacke- 
ray; here evidently no participle), 'no man ever heard of 
opium leading into delirium tremens' (De Quincey), 'the 
suffering arises simply from people not understanding 
this truism' (Ruskin). These examples will show that the 
construction is especially useful in those cases where 
for some reason or other it is impossible to use the 
genitive case, but that it is also found where no such 
reason could be adduced. — Let me sum up by saying that 
when an Englishman now says, 'There is some probability 
of the place having never been inspected by the police', 
he deviates in four points from the constructions of the 

Ing. Gender. 203 

ing that would have been possible to one of his ancestors 
six hundred years ago : place is in the crude form, not in 
the genitive; the adverb; the perfect; and the passive. 
Thanks to these extensions the ing has clearly become a 
most valuable means of expressing tersely and neatly 
relations that must else have been indicated by clumsy 
dependent clauses. 

205. (V. Disappearance of the old word-gender). In 
Old English, as in all the old cognate languages, each 
substantive, no matter whether it referred to animate 
beings or things or abstract notions, belonged to one or 
other of the three gender-classes. Thus he was used in 
speaking of a great many things that had nothing mas- 
culine in their actual nature (e. g. horn^ ende 'end', ehba 
'ebb', dceg 'day') and the feminine pronoun [heo) in regard 
to many which in their nature were not feminine (e. g. 
sorh 'sorrow', glof 'glove', plume 'plum', pipe). Anyone 
acquainted with the intricacies of the same system (or 
want of system) in German will feel how much English 
has gained in clearness and simplicity by giving up these 
distinctions and applying he only to male, and she only 
to female, living beings. The distinction between animate 
and inanimate now is much more accentuated than it 
used to be, and this has led to some other changes, of 
which the two most important are the creation (about 
1600) of the form its (before that time his was neuter as 
well as masculine) and the restriction of the relative 
pronoun which to things : its old use alike for persons and 
things is seen in 'Our father which art in Heaven'. 

206. (VI) A notable feature of the history of the English 
language is the building up of a rich system of tenses on 
the basis of the few possessed by Old English, where the 
present was also a sort of vague future, and where the 

204 VIII. Grammar. 

simple past was often employed as a kind of pluperfect, 
especially when supported by cer 'ere, before'. The use 
of have and had as an auxiliary for the perfect and pluper- 
fect began in the Old English period, but it was then 
only found with transitive verbs, and the real perfect- 
signification had scarcely yet been completely evolved 
from the original meaning of the connexion: ic hcebbe 
l>one fisc gefangenne meant at first 'I have the fish (as) 
caught' (note the accusative ending in the participle). 
By and by a distinction was made between 'I had mended 
the table' and 'I had the table mended', 'he had left 
nothing' and 'he had nothing left'. In Middle English 
have came to be used in the perfect of intransitive verbs 
as well as transitive; / have been does not seem to occur 
earlier than 1200. With such verbs as go and comey I am 
was used in the perfect for several centuries, and / have 
gone and / have come are recent formations. The use of 
will and shall as signs of the future gradually developed 
from the original meaning of 'will' and 'obligation'. The 
periphrastic tenses / am reading, I was reading, I have 
been reading, I shall be reading, were not fully devel- 
oped even in Shakespeare's time ; they are to a great 
extent due to the old construction / am a-reading, where 
a (which afterwards disappeared) represents the prepo- 
sition on and the form in ing is not the participle, but 
the noun. The passive construction (the house is being 
built) is an innovation dating from the end of the eigh- 
teenth century. According to Fitzedward Hall the oldest 
example known is found in a letter from Southey [1795). 
Before that time the phrase was the house is building, i. e. 
is a-building 'is in construction', and the new phrase had 
to fight its way against much violent opposition in the 
nineteenth century before it was universally recognized 
as good English. — While the number of tenses has been 
increased, the number of moods has tended to diminish, 

Tenses. Innovations. 205 

the subjunctive having now very little vital power left. 
Most of its forms have become indistinguishable from 
those of the indicative , but the loss is not a serious 
one, for the thought is just as clearly expressed in 
'if he died', where died may be either indicative or 
subjunctive, as in 'if he were dead', where the verb 
has a distinctively subjunctive form. The verbal system 
has undergone one more important change by the 
extensive use of do as an auxiliary, especially in 
negative and interrogative sentences. This use was not 
regularized in the modern way till the eighteenth century. 

207. (VII) The regularization of the word order (cf. 
§ 14) has been very useful in bringing about clear- 
ness in sentence - construction, and has at the same 
time facihtated many of the simplifications which have 
taken place in the form system and which would other- 
wise have been attended by numerous ambiguities. ^ 

208. (VIII) The pronominal system has been rein- 
forced by some new applications of old material. Who 
and which, originally interrogative pronouns only, are 
now used also as relatives. Self has entered into the 
compounds myself, himself, etc., and has developed a 
plural, ourselves, themselves, which was new in the be- 
ginning of the sixteenth century. With regard to the 
use of these self-iorms it may be remarked that their 
frequency first increased and then in certain cases de- 
creased again: he dressed him, became he dressed himself 
and this is now giving way to he dressed. One has come 
to serve several purposes; as an indefinite pronoun (in 
'one never can tell') it dates from the fifteenth century, 
and as a prop-word ('a little one', 'the little ones') the 

I Cf. Progress in Language p. 89 ff. 

2o6 VIII. Grammar. 

full modern usage goes no further back than to the six- 
teenth century. 

209. (IX) New conjunctions have come into existence 
such as supposing (supposing he comes, what am I to do?), 
provided (I have no objection, provided the benefit is 
mutual), in case (have it ready, in case she should send 
for it, Swift), for jear (they were obliged to drive very 
fast, for fear they should be too late, Dickens), grant 
that (Grant that one has good food ... is that all the 
pay one ought to have for one's work.? Ruskin), like 
(through which they put their heads, like the Guachos 
do through their cloaks, Darwin), directly (Oh ! yes, yes, 
said Kate, directly the whole figure of the singular visitor 
appeared, Dickens), once (once that decision was taken 
his imagination became riotous, H. G. Wells; once you 
are married, there is nothing left for you, not even 
suicide, but to be good, R. L. Stevenson, Virg. Puerisque 
34). It is evident that all these new conjunctions serve to 
vary the modes of joining sentences together and express 
nuances that the old if, when, etc., cannot render in so 
vivid a way; but I am bound to admit that a great many 
Englishmen object to some of them, especially like and once. 

210. (X) The manner in which compound nouns are 
built up has been modified. In compounds of the old 
type the close combination of both nouns is shown by 
the accentual subordination of the second element, cf. 
goldsmith, godson, footstep, leapyear; and very often one 
part, or both, may be phonetically changed, sometimes 
even past recognition, cf. postman, waistcoat, husband, 
hussy (= housewife). But in recent times a new type has 
sprung up in which the second part is not thus accentually 
subordinated to the first, but is stressed at least nearly 
as much as, and sometimes even more than the first 



component. Examples are gold coin, coat tail, village green, 
lead pencil, headmaster.'^ Each part thus is more indepen- 
dent of the other than in the old type, and as an adjective 
is now just as uninflected as a noun forming the first 
part of a compound, the combinations adjective + noun 
and noun + noun are felt to be nearly equivalent. This 
has in recent times led to some curious consequences, 
some examples of which may be here given. We see 
coordination with a true adjective in 'the sepulcher Hath 
op'd his ponderous and marble jawes' (Hamlet), 'with 
thin and rainbow wings' (Tennyson), and still more in 
'home and foreign affairs', 'on some Cumberland or other 
affair' (Carlyle), and in 'a school Latin dictionary', 'an 
evening radical paper'. The use of the prop- word one is 
interesting: 'This umbrella, said Mr. L., producing a fat 
green cotton one' (Dickens), 'most of the mountain flowers 
being lovelier than the lowland ones' (Ruskin). So is the 
use of a qualifying adverb in 'from a too exclusively 
London standpoint', 'in purely Government work' (Lecky), 
'the most everyday occurrences' (Dobson). Thus nouns 
in composition are assuming more and more of the prop- 
erties of the adjectives, and some, as a matter of fact, have 
already become adjectives so completely that they are 
recognized as such by all grammarians : bridal (originally 
brid-ealu 'bride -ale') and dainty (Old French daintie 'a deli- 
cacy', from Latin dignitatem), both assisted by their see- 
mingly adjectival endings, further cheap, chief, choice, etc. 

211. (XI) There are some important innovations in 
the syntax of the infinitive. In such a sentence as 'it is 
good for a man not to touch a woman', the noun with for 
was originally in the closest connexion with the adjective: 

I Cf. on the unstable equilibrium of such compounds my 
Modem Eng. Grammar I p. i54ff. 

2o8 Mil. Grammar. 

'What is good for a man?' 'Not to touch a woman'. But 
by a natural shifting this came to be apprehended as 'it 
is good I for a man not to touch a woman', so that for 
a man was felt to be the subject of the infinitive, and this 
manner of indicating the subject gradually came to be 
employed where the original construction is excluded. 
Thus in the beginning of a sentence: 'For us to levy 
power Proportionate to th'enemy, is all impossible' 
(Shakespeare), and after than: 'I don't know, what is 
worse than for such wicked strumpets to lay their sins at 
honest men's doors' (Fielding); further 'What I like best, 
is for a nobleman to marry a miller's daughter. And what 
I like next best, is for a poor fellow to run away with a 
rich girl' (Thackeray), 'it is of great use to healthy women 
for them to cycle'. ^ Another recent innovation is the use 
of to as what might be called a pro-infinitive instead of 
the clumsy to do so: 'Will you play?' 'Yes, I intend to'. 
'I am going to'. This is one among several indications that 
the linguistic instinct now takes to to belong to the pre- 
ceding verb rather than to the infinitive, a fact which, 
together with other circumstances, serves to explain the 
phenomenon usually mistermed 'the split infinitive'. 
This name is bad because we have many infinitives 
without to, as 'I made him go'. To therefore is no more 
an essential part of an infinitive than the definite article 
is an essential part of a nominative, and no one would 
think of calling 'the good man' a split nominative. Al- 
though examples of an adverb between to and the in- 
finitive occur as early as the fourteenth century, they do 
not become very frequent till the latter half of the nine- 
teenth century. In some cases they decidedly contribute 
to the clearness of the sentence by showing at once what 
word is qualified by the adverb. Thackeray's and Seeley's 

I See my article in Festschrift Vii'tor (Marburg- 1910), p. 85ff. 



sentences 'she only wanted a pipe in her mouth con- 
siderably to resemble the late Field Marshal' and 'the 
poverty of the nation did not allow them successfully to 
compete with the other nations' are not very happily built 
up, for the reader at the first glance is inclined to connect 
the adverb with what precedes. The sentences would 
have been clearer if the authors had ventured to place to 
before the adverb, as Burns does in 'Who dar'd to nobly 
stem tyrannic pride', and Carlyle in 'new Emissaries are 
trained, with new tactics, to, if possible, entrap him, 
and hoodwink and handcuff him'. 

213. This rapid sketch of grammatical changes, though 
necessarily giving only a fraction of the material on which 
it is based, has yet, I hope, been sufficiently full to show 
that such changes are continually going on and that it 
would be a gross error to suppose that any deviation from 
the established rules of grammar is necessarily a corrup- 
tion. Those teachers who know least of the age, origin, 
and development of the rules they follow, are generally 
the most apt to think that whatsoever is more than these 
cometh of evil, while he who has patiently studied the 
history of the past and trained himself to hear the lin- 
guistic grass grow in the present age will generally be 
more inclined to see in the processes of human speech a 
wise natural selection, through which while nearly all 
innovations of questionable value disappear pretty soon, 
the fittest survive and make human speech ever more 
varied and flexible and yet ever more easy and convenient 
to the speakers. There is no reason to suppose that this 
development has come to a stop with the close of the 
nineteenth century: let us hope that in the future the 
more and more almighty schoolmaster may not nip too 
many beneficial changes in the bud. 

Jespersen: English. 2n(i ed. 1 4 

Chapter IX. 
Shakespeare and the Language of Poetry. 

213. In this chapter I shall endeavour to characterize 
the language of the greatest master of English poetry 
and make some observations in regard to his influence 
on the English language as well as in regard to poetic 
and archaic language generally. But it must be distinctly 
understood that I shall concern myself with language and ij 
not with literary style. It is true that the two things 
cannot be completely kept apart, but as far as possible 1 
shall deal only with what are really philological as opposed 
to literary problems. 

214. Shakespeare's vocabulary is often stated to be 
the richest ever employed by any single man. It has been 
calculated to comprise 2i,000 words ('rough calculation, 
found in Mrs. Clarke's Concordance . . . without counting 
inflected forms as distinct words', Craik), or, according 
to others, 24,000 or 15,000. In order to appreciate what I 
that means we must look a little at the various statements 
that have been given of the number of words used 
by other authors and by ordinary beings, educated and 
not educated. Unfortunately these statements are in 
many cases given and repeated without any indication 
of the manner in which they have been arrived at.^ Mil- 

I Max Miiller, IVissenschaft der Sprache I 360 and Lectures 
on the Science of Language , 6th ed. I 309. Elze, William Shake- 
speare, Halle 1876, 449, Wundt, Volkerpsychologie, Sprache II, 

Vocabulary. 2 1 1 

ton's vocabulary is said to comprise 7000 or 8000 words, 
that of the Iliad and Odyssey taken together 9000, that 
of the Old Testament 5642 and that of the New Testa- 
ment 4800. 

215. Max Miiller says that a farm-labourer uses only 
300 words, and Wood that 'the average man uses about 
five hundred words' (adding 'it is appalling to think how 
pitiably we have degenerated from the copiousness of 
our ancestors'). But both figures are obviously wrong. 
One two-year-old girl had 489 and another 1121 words 
(see Wundt), while Mrs. Winfield S. Hall's boy used in 
his 17th month 232 different words and, when six years 
old, 2688 words — at least, for it is probable that the 
mother and her assistants who noted down every word 
they heard <he child use, even so did not get hold of its 
whole vocabulary. Now, are we really to believe, with 
Wundt, that the linguistic range of a grown-up man, 
however humble, is considerably smaller than that of a 
two-year-old child of educated parents or is only one- 
seventh of that of a six-year-old boy! Any one going 
through the lists given by Mrs. Hall will feel quite certain 
that no labourer contents himself with so scanty a vocab- 
ulary. Schoolbooks for teaching foreign languages often 
include some 700 words in the first year's course; yet on 
how few subjects of everyday occurrence are our pupils 

Leipz. 1900, 308. Wood, Journal of Germanic Philology I 294. 
Craik, Engl. Language and Literature 264. Emerson, History 
of the Engl. Language, 1894, 114. Le Maitre Phonetique 1888, 
47. Smedberg , ^S'2/^«J/^« landsmalen Xi , 9 (S7) 1896. Marius 
Kristensen, Aarbog for dansk kulturhistorie 1897. Babbitt, 
Common Sense in Teachiftg Modern Languages, New York 1895, 
II. Svi&ti, History of Language, 1900, iZ9- Weise, Utisere 
Muttersprache , 1897, 205, Dewischeit, Shakespeare -fahrbuch 
XXXIV (1898) 190. Mrs. Winfield S. Hall , Child Study , Monthly, 
March 1897 2ind Journal of Childhood and Adolescence, January 


2 12 IX. Shakespeare and the Language of Poetry. 

able to converse after one year's teaching. Sweet also 
contradicts the statement about 300 words, saying 'When 
we find a missionary in Tierra del Fuego compiling a 
dictionary of 30,000 words in the Yaagan language — 
that is, a hundred times as many — we cannot give any 
credence to this statement, especially if we consider the 
number of names of different parts of a waggon or a 
plough, and all the words required in connexion even 
with a single agricultural operation, together with names 
of birds, plants, and other natural objects'. Smedberg, 
who has investigated the vocabulary of Swedish peasants 
and who emphasizes its richness in technical teims, 
arrives at the result that 26,000 is probably too small a 
figure, and the Danish dialectologist Kristensen com- 
pletely endorses this view. Professor E. S. Holden tested 
himself by a reference to all the words in Webster's 
Dictionary, and found that his own vocabulary com- 
prised 33,456 words. And E. H. Babbitt writes: 'I tried 
to get at the vocabulary of adults and made experiments, 
chiefly with my students, to see how many English words 
each knew . . . My plan was to take a considerable number 
of pages from the dictionary at random, count the number 
of words on those pages which the subject of the ex- 
periment could define without any context, and work 
out a proportion to get an approximation of the entire 
number of words in the dictionary known. The results 
were surprising for two reasons. In the size of the vocab- 
ulary of such students the outside vaiiations were les-^. 
than 20 per cent., and their vocabulary was much larger 
than I had expected to find. The majority reported a 
little below 60,000 words'. 

216. These statements are easily reconciled with the 
ascription of 20,000 words to Shakespeare. For it must 
be remembered that in the case of each of us there is a 
great difference between the words known (especially 

Vocabulary. 213 

those of which he has a reading knowledge) and the words 
actually used in conversation. And then, there must 
always be a great many words which a man will use 
readily in conversation, but which will never occur in 
his writings, simply because the subjects on which a man 
addresses the public are generally much less varied than 
those he has to talk about every day.^ How many authors 
have occasion to use in their books even the most familiar 
names of garden tools or common dishes or kitchen 
implements? When Milton as a poet uses only 8,000 
against Shakespeare's 20,000 words, this is a natural 
consequence of the narrower range of his subjects, 
and it is easy to prove that his vocabulary really 
contained many more than the 8,000 words found in 
a Concordance to his poetical works. We have only 
to take any page of his prose writings, and we 
shall meet with a great many words not in the Con- 

217. The greatness of Shakespeare's mind is therefore 
not shown by the fact that he was acquainted with 20,000 
words, but by the fact that he wrote about so great a 
variety of subjects and touched upon so many human 
facts and relations that he needed this number of words 

1 Inversely, many authors will use some (learned or abstract) 
words in writing which they do not use in conversation; their 
number, however, is rarely great. 

2 Thus, on p. 30 of Areopagitica I find the following 21 
words, which are not in Bradshaw's Concordance: churchman, 
competency, utterly, mercenary, pretender, ingenuous, evi- 
dently, tutor, examiner, seism, ferular, fescu, imprimatur, 
grammar, pedagogue, cursory, temporize, extemporize, licencer, 
commonwealth, foreiner. And p. 50 adds 18 more words to 
the list: writing, commons, valorous, rarify, enfranchise, 
founder, formall, slavish, oppressive, reinforce, abrogate, 
mercilesse, noble (n.), Danegelt, immunity, newnes, unsut- 
ablenes, customary. 

2 14 IX. Shakespeare and the Language of Poetry. 

in his writings.^ His remarkable familiarity with technical 
expressions in many different spheres has often been 
noticed, but there are other facts with regard to his use 
of words that have not been remarked, or not suf- 
ficiently remarked. His reticence about religious mat- 
ters, which has given rise to the most divergent 
theories of his rehgious behef, is shown strikingly in 
the fact that such words as Bible, Holy Ghost, and 
Trinity do not occur at all in his writings, while Jesus 
(Jesu) , Christ and Christmas are found only in some of 
his earliest plays; Saviour occurs only once (in Hamlet), 
and Creator only in two of the dubious plays (H 6 C and 

218. Of far greater importance is his use of language 
to individualize the characters in his plays. In this he 
shows a much finer and subtler art than some modern 
novelists, who make the same person continually use the 
same stock phrase or phrases. Even where he resorts to 
the same tricks as other authors he varies them more; 
Mrs. Quickly and Dogberry do not misapply words from 
the classical languages in the same way. The everyday 
speech of the artisans in A Midsummer Night's Dream 
is comic in a different manner from the diction they use 
in their comedy, which serves Shakespeare to ridicule 
some linguistic artifices employed in good faith by many 
of his contemporaries (alliteration, bombast). Shake- 
speare is not entirely exempt from the fashionable affec- 

1 I have amused myself with making up the following sen- 
tences of words not used by Shakespeare though found in the 
language of that time: In Shakespeare we find no blunders, 
although decency and delicacy have disappeared; energy and 
enthusiasm are not in existence , and we see no elegant express- 
ions nor any gleams oi genius , etc. 

2 The act against profane language on the stage (see below, 
% 244) is not sufficient to explain this reticence. 

Individual Characters, 


tation of his days known as Euphuism^, but it must be 
noticed that he is superior to its worst aberrations and 
he satirizes them, not only in Love's Labour's Lost, but 
also in many other places. Euphuistic expressions are 
generally put in the mouth of some subordinate character 
who has nothing to do except to announce some trifling 
incident, relate a little of the circumstances that lead up 
to the action of the play, deliver a message from a king, 
etc. It is not improbable that the company possessed 
some actor who knew how to make small parts funny by 
imitating fashionable affectation, and we can imagine 
that it was he who acted Osric in Hamlet, and by his 
vocabulary and appearance exposed himself to the scoffs 
of the Danish prince, the Captain in Twelfth Night I, sc. 2, 
the Second Gentleman in Othello II, sc. i, the first Lord 
in As You Like It II, sc. 2 (They found the bed un- 
treasur'd of their mistris'). But the messenger from 
Antony in Julius CcBsar (III. i. 122) speaks in a totally 
different strain and gives us a sort of foretaste of Antony's 
eloquence. And how different again — I am speaking 
here of subordinate parts only — are the gardeners in 
Richard the Second (III, sc. 4) with their characteristic 
application of botanical similes to politics and vice versa. 
And thus one might go on, for no author has shown 
greater skill in adapting language to character. 

219. A modern reader, however, is sure to miss many 
of the nuances that were felt instinctively by the poet's 
contemporaries. A great many words have now another 
value than they had then; in some cases it is only a 
slightly different colouring, but in others the diversity is 
greater, and only a close study of Elizabethan usage can 

I The various kinds of affected court style have been care- 
fully distinguished by M. Basse, Stijlaffectatie bij Shakespeare, 
voorall uit het oogpunt van het Euphuisme (University de Gand 


2 1 6 IX. Shakespeare and the Language of Poetry. 

bring out the exact value of each word. A bonnet then 
meant a man's cap or hat; Lear walks unbonneted. To 
charm always implied magic power, to make invulnerable 
by witchcraft, to call forth by spells etc.; 'charming 
words' were magic words and not simply delightful words 
as in our days. Notorious might be used in a good sense 
as 'well-known'; censure^ too, was a colourless word ('And 
your name is great In mouthes of wisest censure' 0th. II, 
3. 193). The same is true of succeed and success, which 
now imply what Shakespeare several times calls 'good 
success', whereas he also knows 'bad success'; cf. 'the 
effects he writes of succeede unhappily' Lear I. 2. 157. 
Companion was often used in a bad sense, lik-e fellow now, 
and inversely sheer, which is now used with such words 
as 'folly, nonsense', had kept the original meaning of 
'pure', as in 'thou sheere, immaculate, and silver foun- 
taine' (R 2 V. 3. 61). Politician seems always to imply 
intriguing or scheming, and remorse generally means pity 
or sympathy. Accommodate evidently did not belong to 
ordinary language, but was considered affected; occupy 
and activity were at least half-vulgar, while on the other 
hand wag (vb.) was then free from its present trivial or 
ludicrous associations ('Untill my eielids will no longer 
wag', Hamlet V. i. 290, see Dowden's note on this pas- 
sage). Assassination (only Macbeth I. 7. 2) would then 
call up the memory of the 'Assasines, a company of most 
desperat and dangerous men among the Mahometans' 
(Knolles, Hist. Turks 1603) or 'That bloudy sect of 
Sarazens, called Assassini, who, without feare of torments, 
undertake . . . the murther of any eminent Prince, im- 
pugning their irreligion' (Speed, 161 1, quoted N. E. D.) 
220. Even adverbs might then have another colouring 
from their present signification. Now-a-days was a vulgar 
word; it is used by no one in Shakespeare except Bottom, 
the grave-digger in Hamlet, and a fisherman in Pericles. 

Value of Words, Bacon. 


The adverb eke, in the nineteenth century a poetic word, 
seems to have been a comic expression; it occurs only- 
three times in Shakespeare (twice in the Merry Wives, 
used by Pistol and the Host, once by Flute in Mids. 
N. Dr.); Milton and Pope avoid the word. The synonym 
also is worth noticing. Shakespeare uses it only 22 times, 
and nearly always puts it in the mouth of vulgar or 
affected persons (Dogberry twice in Ado, the Clown once 
in Wint., the Second Lord in As II. sc. 2, the Second Lord 
in Tim. III. sc. 6, the affected Captain in Tw. I. sc. 2; 
the knight in Lear I. 4. 66 may belong here too ; further 
Pistol twice in grandiloquent speeches, H 4 B II. 4. 171 
and V. 3. 145, and two of Shakespeare's Welshmen, 
Evans three times, and Fluellen twice). It is used twice 
in solemn and official speeches (H 5 I. 2. yy, where Canter- 
bury expounds lex Salica, and IV. 6. 10), and it is, there- 
fore, highly characteristic that Falstaff uses the word 
twice in his Euphuistic impersonation of the king (H 4 
A II. 4. 440 and 459) and twice in similar speeches in the 
Merry Wives (V. i. 24 and V. 5. 7).^ 

I The only passages not accounted for above are Gent. III. 
2. 25, where the metre is wrong, Hamlet V. 2. 402, where the 
folios have always instead of also, and Caes. II. i. 329. — Shake- 
speare's sparing use of also would in itself suffice to disprove | \ 
the Baconian theory if any proof were needed beyond the evi- 
dence of history and of psychology. For in Bacon, alsds abound, 
and I have counted on four successive small pages of Moore 
Smith's edition of the New Atlantis 22 instances, exactly as 
many as are found in the whole of Shakespeare. Might and 
^nought seem to be nearly equally frequent in Bacon, but mought 
is found only once in Shakespeare , in the third part of Henry 
VI, a play which many competent judges are inclined not to 
ascribe to Shakespeare at all. At any rate, this one instance 
in one of his earliest works weighs nothing as against the thou- 
sands of times might is found. Shakespeare uses among and 
amoitgst indiscriminately. Bacon nearly always uses amongst. 
aeon frequently employs the conjunction whereas, which is 
not found at all in the undoubtedly genuine Shakespearian 

K sa 

2 1 8 IX. Shakespeare and the Language of Poetry. 

221. Shylock is one of Shakespeare's most interesting 
creations, even from the point of view of language. Al- 
though Sidney Lee has shown that there were Jews in 
England in those times and that, consequently, Shake- 
speare need not have gone outside his own country in 
order to see models for Shylock, the number of Jews 
cannot have been sufficient for his hearers to be very 
familiar with the Jewish type, and no Anglo-Jewish dia- 
lect or mode of speech had developed which Shakespeare 
could put into Shylock's mouth and so make him at once 
recognizable for what he was. I have not, indeed, been 
able to discover a single trait in Shylock's language that 
can be called distinctly Jewish. And yet Shakespeare has 
succeeded in creating for Shylock a language different 
from that of anybody else. Shylock has his Old Testa- 
ment at his fingers' ends, he defends his own way of 
making money breed by a reference to Jacob's thrift in 
breeding parti-coloured lambs, he swears by Jacob's staff 
and by our holy sabbath, and he calls Lancelot 'that 
foole of Hagars off-spring'.^ We have an interesting bit 
of Jewish figurative language in 'my houses eares, I 
meane my casements' (II. 5. 34). Shylock uses some 
biblical words which do not occur elsewhere in Shake- 
speare: pilled (The skilful shepheard pil'd me certain 
wands, cf. Genesis XXX. 37), synagogue, Nazarite, and 
publican. But more often Shylock is characterized by 
being made to use words or constructions a little different 
from the accepted use of Shakespeare's time.^ He dis- 
plays, etc. — Since this was written, the whole subject has 
been investigated by N, Begholm {Bacon og Shakespeare, Copen- 
hagen 1906), who has succeeded in pointing out an astonishing 
number of discrepancies between the two authors. 

1 Contrast with this trait the fondness for classical allusions 
found in Marlowe's Barrabas. 

2 He says Abra?n, but Abraham is the only form found in 
the rest of Shakespeare's works. 

Shylock. 2 [ 9 

likes the word interest and prefers calling it advantage 
or thrift (my well-worne thrift, which he cals interrest, 
I. 3. 52), and instead of usury he says usance. Furness 
quotes Wylson On Usurye 1572, p. 32 'usurie and double 
usurie, the merchants termyng it usance and double 
usance, by a more clenlie name' — this word thus ranks 
in the same category as dashed or d-d for damned: instead 
of pronouncing an objectionable word in full one begins 
as if one were about to pronounce it and then shunts off 
on another track (see other examples below, § 244). 
Shylock uses the plural moneys, which is very rare in 
Shakespeare, he says an equal pound for 'exact', rheum 
(rume) for 'saliva', estimable for 'valuable', fulsome for 
'rank' (the only instance of that signification discovered 
by the editors of the N. E. D.) ; he alone uses the words 
eaneling and misbeliever and the rare verb to bane. His 
syntax is peculiar: we trifle time; rend out, where Shake- 
speare has elsewhere only rend; I have no mind of feasting 
forth to-night (always mind to); and so following, where 
and so forth is the regular Shakespearian phrase. I have 
counted some forty such deviations from Shakespeare's 
ordinary language and cannot dismiss the thought that 
Shakespeare made Shylock's language peculiar on pur- 
pose, just as he makes Caliban, and the witches in Macbeth, 
use certain words and expressions used by none other 
of his characters in order to stamp them as beings out 
of the common sort. 

222. Shakespeare's vocabulary was not the same in 
all periods of his life. I have counted between two and 
three hundred words which he used in his youth, but not 
later, while the number of words peculiar to his last period 
is much smaller. Sarrazin^ mentions as characteristic of 
his first period a predilection for picturesque adjectives 

I Shakespeare- J ahrbuch XXXIII, 122. 

2 20 1^- Shakespeare and the Language of Poetry. 

that appeal immediately to the outward senses (bright, 
brittle, fragrant, pitchy, snow-white), while his later plays 
are said to contain more adjectives of psychological im- 
portance. But even apart from the fact that some of the 
adjectives instanced are really found in later plays {bright 
in Caes., Ant., 0th., Cymb., Wint. T., etc.), this statement 
would account for only a small part of the divergencies. 
Probably no single explanation can account for them all, 
not even that of the natural buoyancy of youth and the 
comparative austerity of a later age. It is noteworthy 
that in some instances he ridicules in later plays words 
used quite seriously in earlier ones. Thus beautify, which 
is found in Lucrece, Henry VI B, Titus Andr., Two 
Gentlemen, and Romeo, is severely criticized by Polonius 
when he hears it in Hamlet's letter : 'That's an ill phrase, 
a vilde phrase, beautified is a vilde phrase'. Similarly 
cranny, which Shakespeare used in Lucrece (twice) and 
in the Comedy of Errors, is not found in any play written 
later than Mids N. D., where Shakespeare takes leave of 
the word by turning it to ridicule in the mouth of Bottom 
and in the artisans' comedy. The fate of foeman, aggra- 
vate, and homicide is nearly the same. Perhaps some of 
the words avoided in later life were provinciahsms (thus 
possibly pebblestone, shore in the sense of 'bank of a river', 
wood 'mad', forefather 'ancestor', the pronunciation of 
marriage and of Henry in three syllables). In the first 
period Shakespeare used perverse with the unusual signi- 
fication 'cold, unfriendly, averse to love', later he avoids 
the word altogether. In such instances he may have 
been criticized by his contemporaries (we know from the 
Poetaster how severe Ben Jonson was in these matters), 
and that may have made him avoid the objectionable 
words altogether. 

223. One of the most characteristic features of Shake- 
speare's use of the English language is his boldness. 

Different Periods. Boldness. 2 2 i 

His boldj i^s ,? <^^ mftf:aphor ha s often been pointed out in 
books of literary criticism, and the boldness of his sen- 
t ence structure , especially in his last period, is so obvious 
that no instances need be adduced here. He does not 
always care for grammatical parallelism, witness such a 
sentence as 'A thought which, quarter'd, hath but one 
part wisedom And ever three parts coward' (Haml. IV. 
4. 42). He does not always place the words where they 
would seem properly to belong, as in 'we send. To know 
what willing ransome he will give' for 'what ransom he 
will willingly give' (H 5 HI. 5. 63), 'dismist me Thus with 
his speechlesse hand' (Cor. V. I. 68), 'the whole eare of 
Denmarke Is by a forged processe of my death Rankly 
abus'd' (the ear of all Denmark, Haml. I. 5. 36), 'lovers 
absent howres' (the hours when lovers are absent, 0th. 
III. 4, 174) etc. He is not afraid of writing 'wanted lesse 
impudence' for 'had less impudence' or 'wanted impu- 
dence more' (Wint. III. 2. 57) and 'a begger without lesse 
quality' (Cymb. I. 4. 23), nor of mixing his negatives as 
he does in many other passages.^ Al. Schmidt, who 
collects many instances of such negligence, rightly re- 
. marks: 'Had he taken the pains of revising and preparing 
I his plays for the press, he would perhaps have corrected 
all the quoted passages. But he did not write them to be 
read and dwelt on by the eye, but to be heard by a sym- 
pathetic audience. And much that would blemish the 
language of a logician, may well become a dramatic poet 
or an orator'. ^ There is an excellent paper by C. Alphonso 
Smith in the Englische Studien, vol. XXX, on 'The Chief 
Difference between the First and Second Folios of Shake- 
speare', in which he shows that 'the supreme syntactic 
value of Shakespeare's work as represented in the First 

1 Besides using such double negatives as were regular in all 
the older periods of the language {nor never, etc.) 

2 Shakespeare -Lexicon, p. 1420. 

2 22 IX. Shakespeare and the Language of Poetry. 

Folio is that it shows us the English language unfettered 
/ by bookish impositions. Shakespeare's syntax was that 
j of the speaker, not that of the essayist; for the drama 
\ represents the unstudied utterance of people under all 
|! kinds and degrees of emotion, ennui, pain, and passion. 
I Its syntax, to be truly representative, must be familiar, 
(^conversational, spontaneous; not studied and formal.' 
But 'the Second Folio is of unique service and significance 
in its attempts to render more 'correct' and bookish the 
unfettered syntax of the First. The First Folio is to the 
Second as spoken language is to written language'. The 
'bad grammar' of the First Folio (1623) may not always 
be due to Shakespeare himself, but at any rate we have in 
that edition more of his own language than in the 'cor- 
^ rectness' of the Second Folio (1632). 

224. Shakespeare's boldness with regard to language 
is less conspicuous, though no less real, in the instances 
I shall now mention. In turning over the pages of the 
New English Dictionary, where every pains has been 
taken to ascertain the earliest occurrence of each word 
and of each signification, one is struck by the frequency 
with which Shakespeare's name is found affixed to the 
earliest quotation for words or meanings. In many cases 
this is no doubt due to the fact that Shakespeare's vocab- 
ulary has been registered with greater care in Concord- 
ances and in Al. Schmidt's invaluable Shakespeare- 
Lexicon than that of any other author, so that his w'ords 
cannot escape notice, while the same words may occur 
unnoticed in the pages of many an earlier author. But 
even if future research may somewhat reduce the number 
of these words, the fact will remain that Shakespeare 
was in no way afraid of adopting into his immortal 
pages a great many words which were new in his times, 
whether absolutely new or new only to the written lan- 
guage, while living colloquially on the lips of the people. \ 


New Words. 


My list includes the following words: aslant as a pre- 
position, assassination (see above), barefaced, beguile in 
two of the significations now most current (win the at- 
tention by wiling means, and charm away), the plural 
brothers (found also in Layamon's Brut, but seemingly 
not between that and Shakespeare's Titus Andron. and 
Marlowe's T amburlaine) , call 'to pay a short visit', court- 
ship, dwindle, enthrone (earlier enthronize), eventful, ex- 
cellent in the current sense 'extremely good', fount 
'spring', fretful, get intransitive with an adjective, 'be- 
come' (only in 'get clear'), / have got for 'I have', gust, 
hint, hurry, indistinguishable, latest, laughable, leap-frog, 
loggerhead and loggerheaded, lonely (but Sidney has lone- 
liness some years before Shakespeare began to write), 
lower verb, perusal, primy. Further the following verbs 
(formed from nouns that are found before Shakespeare's 
time) : bound, hand, jade, and nouns (formed from already 
existing verbs) : control, dawn, dress, hatch, import, indent. 
Among other words which were certainly or probably 
new when Shakespeare used them, may be mentioned 
acceptance, gull 'dupe', rely, scarcely, and summit. I shall 
give below (§ 228) a list of words and expressions the 
existence of which in the English language is due to Shake- 
speare. The words here given would probably have found 
their way mto the language even had Shakespeare never 
written a line, though he may have accelerated the date 
of their acceptance. But at any rate they show that he 
was exempt from that narrowness which often makes 
authors shy of using new or colloquial words in the higher 
literary style. Let me add another remark apropos of a 
list of hard words needing an explanation which is found 
in Cockeram's Dictionarie (1623). Dr. Murray writes^: 
*We are surprised to find among these hard words abandon, 

I The Evolutio7i of English Lexicography. Romanes Lecture, 
Oxford and London 1900, p. 29. 

2 24 ^^' Shakespeare and the Language of Poetry, 

abhorre, abrupt, absurd, action, activitie, and actresse, ex- 
plained as 'a woman doer', for the stage actress had not 
yet appeared'. Now, with the exception of the last one, 
all these words are found in Shakespeare's plays. 

225. Closely connected with this trait in Shakespeare's 
language is the proximity of his poetical diction to his 
ordinary prose. He uses very few 'poetical' words or 
forms. He does not rely for his highest flights on the 
use of words and grammatical forms not used elsewhere, 
but knows how to achieve the finest effects of imagination 
without stepping outside his ordinary vocabulary and 
grammar. It must be remembered that when he uses 
thou and thee, 'tis, e'en, ne'er, howe'er, mine eyes, etc., or 
when he construes negative and interrogative verbs 
without do, all these things which are now parts of the 
conventional language of poetry, were everyday collo- 
quialisms in the Elizabethan period. It is true that there 
are certain words and forms which he never uses except 
in poetry, but their number is extremely small. I do not 
know""^of any besides host 'army', vale, sire, and morn. As 
for the synonym morrow, apart from its use in the sense 
of 'next day' and in the salutation good morrow, which 
was then colloquial, it occurs only four'^times, and only 
in rime. There are some verb forms which only occur in 
rime, but the number of occasions on which Shakespeare 
was thus led to deviate from his usual grammar is very 
small: begun (past tense) 8 times, flee once (the usual 
present is fly), gat once (in the probably spurious Pericles), 
sain once, sang once, shore participle once, strown once 
(the usual form is strewed), swore participle once — fifteen 
instances in all, to which must be added eleven instances 
of the plural eyen. Rhythmical reasons seem to make do 
more frequent in Shakespeare's verse than in his prose\ 

I W.Franz, Shakespeare-Grammatik, 2nd ed.478. His statistics 
might be more comprehensive. 

Poetical Diction. 


and rhythm and rime sometimes make him place a pre- 
position after instead of before the noun [e. g. go the fools 
among.^) All these things are rare enough to justify the 
statement that a peculiar poetical diction is practically 
non-existent in Shakespeare. 

226. In the Old English period the language of poetry 
differed, as we have seen (cf. § 53), very considerably 
from the language of ordinary prose. The old poetical 
language was completely forgotten a few centuries after 
the Norman Conquest, and a new one did not develop in 
the Middle English period, though there were certain 
conventional tricks used by many poets, such as those 
ridiculed in Chaucer's Sir Thopas. Chaucer himself had 
not two distinct forms of language, one for verse and the 
other for prose, apart from those unavoidable smaller 
changes which rhythm and rime are always apt to bring 
about. We have now seen that the same is true of Shake- 
speare; but in the nineteenth century we find a great 
many words and forms of words which are scarcely ever 
used outside of poetry. This, then, is not a survival of 
an old state of things, but a comparatively recent phenom- 
enon, whose causes are well worth investigating. At 
first it might be thought that the regard for sonority 
and beauty of sound would be the chief, or one of the 
chief agents in the creation of a special poetical dialect. 
But very often poetical forms are, on the contrary, less 
euphonious than everyday forms; compare for example 
break' st thou with do you break. Those who imagine that 
gat sounds better than got will scarcely admit that spat 
or gnat sounds better than spot or not: non-phonetic 
associations are often more powerful than the mere sounds. 

227. More frequently it is the desire to leave the beaten 
track that leads to the preference of certain words in 

I Franz, p. 427. 

Jespbrsbn: English, and ed. 1 5 

2 26 IX. Shakespeare and the Language of Poetry. 

poetry. Words that are too well known and too often 
used do not call up such vivid images as words less 
familiar. This is one of the reasons which impel poets to 
use archaic words; they are 'new' just on account of their 
being old, and yet they are not so utterly unknown as 
to be unintelligible. Besides they will often call up the 
memory of some old or venerable work in which the 
reader has met with them before, and thus they at once 
secure the reader's sympathy. If, then, the poetical 
language of the nineteenth century contains a great 
many archaisms, the question naturally presents itself, 
from what author or authors do most of them proceed? 
And many people who know the preeminent position of 
Shakespeare in EngHsh literature will probably be sur- 
prised to hear that his is not the greatest influence on 
English poetic diction. 

228. Among words and phrases due to reminiscences 
of Shakespeare may be mentioned the following: antre 
(Keats, Meredith), atomy in the sense 'atom, tiny being', 
beetle (the dreadfull summit of the cliffe. That beetles o'er 
his base into the sea), it beggars all description, broad- 
blown, charactery (Keats, Browning), coign of vantage 
(coign is another spelling of coin 'corner'), cudgel one's 
brain(s), daff the world aside, eager 'cold' (a nipping and 
an eager ayre), eld (superstitious eld), nine farrow, fitful 
(Lifes fitfull fever), forcible feeble, a foregone conclusion, 
forgetive (Falstaff; 'of uncertain formation and meaning. 
Commonly taken as a derivation of forge v., and hence 
used by writers of the 19th c. for: apt at forging, in- 
ventive, creative' N. E. D.), a forthright (rare), gaingiving 
(Coleridge), gouts of blood, gravelblind, head and front 
('A Shakesperian phrase, orig. app. denoting 'summit, 
height, highest extent or pitch'; sometimes used by 
modern writers in other senses'. N. E. D.), hoist with 
his own petard, lush (in the sense 'luxuriant in growth'), 

Words from Shakespeare. 2 27 

in my mind's eye, the pink (of perfection, in Shakespeare 
only 'I am the very pinck of curtesie'; George Eliot has 
'Her kitchen always looked the pink of cleanliness', and 
Stevenson 'he had been the pink of good behaviour'), 
silken dalliance, single blessedness, that way madness lies 
('Too kind ! Insipidity lay that way', Mrs. Humphrey 
Ward), weird. The last word is interesting; originally it 
is a noun and means 'destiny, fate'; the three weird 
sisters means the fate sisters or Norns. Shakespeare found 
this expression in Holinshed and used it in speaking of 
the witches in Macbeth, and only there. From that play 
it entered into the ordinary language, but without being 
properly understood. It is now used as an adjective and 
generally taken to mean 'mystic, mysterious, unearthly'. 
Another word that is often misunderstood is bourne from 
Hamlet (The undiscovered countrey, from whose borne 
No traveller returnes); it means 'limit', but Keats and 
others use it in the sense 'realm, domain' (In water, fiery 
realm, and airy bourne; quoted N. E. D.). There are two 
things worth noting in this list. First, that it includes so 
many words of vague or indefinite meaning, which were 
not perhaps even clearly understood by the author him- 
self. This explains the fact that some of them have 
apparently been used in modern times in a different sense 
from that intended by Shakespeare. Second, that the 
re-employment of these words nearly always dates from 
the nineteenth century and that the present currency 
of some of them is due just as much to Sir Walter Scott 
or Keats as to the original author. To cudgel one's brains 
is now more of a literary phrase than when Shakespeare 
put it in the mouth of the gravedigger (Hamlet V. I. 63), 
evidently meaning it to be a rude or vulgar expression. 
Inversely, single blessedness is now generally used with 
an ironical or humorous tinge which it certainly had not 
in Shakespeare (Mids. I. i. 78). 


2 28 IX. Shakespeare and the Language of Poetry, 

229. It must be noted also that none of the words thus 
traceable to Shakespeare belong now to what might be 
called the technical language of poetry. Modern archaiz- 
ing poetry owes its vocabulary more to Edmund Spenser 
than to any other poet. Pope and his contemporaries 
made a very sparing use of archaisms, but when poets 
in the middle of the eighteenth century turned from his 
rationalistic and matter-of-fact poetry and were eager to 
take their romantic flight away from everyday realities, 
Spenser became the poet of their heart, and they adopted 
a great many of his words which had long been forgotten. 
Their success was so great that many words which they 
had to explain to their readers are now perfectly familiar 
to every educated man and woman. Gilbert West, in his 
work 'On the Abuse of Travelling, in imitation of Spenser' 
(1739) had to explain in footnotes such words as sooth, 
guise, hardiment, Elfin, prowess, wend, hight, dight, para- 
mours, behests, caitiffs'^. William Thompson, in his 'Hymn 
to May' (1740?) explains certes surely, certainly, ne nor, 
erst formerly, long ago, undaz'd undazzled, sheen biight- 
ness, shining, been are, dispredden spread, meed prize, ne 
recks nor is concerned, affray affright, featly nimbly, 
defftly finely, glenne a country borough, eld old age, lusty- 
head vigour, algate ever, harrow destroy, carl clown, perdie 
an old word for asserting anything, livelood liveliness, 
albe altho', scant scarcely, bedight adorned. 

230. In later times, Coleridge, Scott, Keats, Tennyson, 
William Morris, and Swinburne must be mentioned as 
those poets who have contributed most to the revival of 
old words. Coleridge in the first edition of the Ancient 
Mariner used so many archaisms in spelling, etc., that 
he had afterwards to reduce the number in order to make 
his poem more palatable to the reading public. Some- 

I W. L. Phelps, Beginnings of the Romantic Movement, p. 63. 

Archaisms. 229 

times pseudo-2intique formations have been introduced; 
anigh, for instance, which is frequent in Morris, is not an 
old word, and idlesse is a false formation after the legiti- 
mate old noblesse and humblesse (O. Fr, noblesse, hum- 
blesse). But on the whole, many good words have been 
recovered from oblivion, and some of them will 
doubtless find their way into the language of ordinary 
conversation, while others will continue their life in 
the regions of higher poetry and eloquence. On the 
other hand , many pages in the works of Shakespeare, 
of Shelley, and of Tennyson show us that it is possible 
for a poet to reach the highest flights of eloquent 
poetry without resorting to many of the conventionally 
poetical terms. 

231. As for the technical grammar of modern poetry, 
the influence of Shakespeare is not very strong, in fact 
not so strong as that of the Authorized Version of the 
Bible. The revival of th in the third person singular was 
due to the Bible, as we have seen above (§ 196)^. Gat is 
a frequent form in the Bible, while Shakespeare's ordinary 
past of the verb to get is got; the solitary instance of gat 
(see § 225) only serves to confirm the rule^. The past 
tense of cleave 'to sever' in Shakespeare is clove or cleft; 
clave does not occur in his writings at all, but is the only 
biblical past of this verb. Brake is the only preterite of 

1 When modern clergymen in reading the Bible pronounce 
laved, danced, etc., they are reproducing a language about two 
hundred years earlier than the Authorized Version. 

2 Gai is the only form of this verb admitted by some modern 
poets, who avoid get and got altogether. Shakespeare uses the 
verb hundreds of times. Milton makes a very sparing use of 
the verb (which he inflects get got got, never gat in the past 
or gotten in the participle); all the forms of the verb only occur 
1 9 times in his poetical works, while, for instance, give occurs 
168 times and receive 73 times. The verb is rare in Pope too. 
Why is this verb tabooed in this way? 

230 IX. Shakespeare and the Language of Poetry. 

break found in the Bible; in Shakespeare brake is rarer 
than broke; Milton and Pope have only broke; Tennyson, 
Morris, and Swinburne prefer brake. 
/i 232. But on the whole, modern poets do not take their 
grammar from any one old author or book, but are apt 
to use any deviation from the ordinary grammar they can 
lay hold of anywhere. And thus it has come to pass in 
the nineteenth century that while the languages of other 
civilized nations have the same grammar for poetry as 
for prose, although retaining here and there a few archaic 
forms of verbs, etc., in English a wide gulf separates the 
grammar of poetry from that of ordinary life. The pro- 
noun for the second person is in prose you for both cases 
in both numbers, while in many works of poetry^it is thou 
and thee for the singular, ye for the plural (with here and 
there a rare you)] the poetical possessives thy and thine 
never occur in everyday speech. The usual distinction 
between my and mine does not always obtain in poetry, 
where it is thought refined to write mine ears, etc. For 
they sat down the poetical form is they sate them down; for 
it's poets write 'tis, and for whatever either whatso or 
whatsoever (or whate'er), for does not mend they often 
write mends not, etc. Sometimes they gain the advantage 
of having at will one syllable more or less than common 
people : taketh for takes, thou takest for you take, movkd for 
moved, o'er for over, etc.; compare also morn for morning. 
But in other cases the only thing gained is the impression, 
produced by uncommon forms, that we are in a sphere 
different from or raised above ordinary realities. As a 
matter of course, this impression is weakened in pro- 
portion as the deviations become the common property 
of any rimer, when a reaction will probably set in in 
favour of more natural forms. The history of some of the 
poetical forms is rather curious : howe'er, e'er, o'er, e'en were 
at first vulgar or familiar forms, used in daily talk. Then 

Grammar of Poetry. 2 3 I 

poets began to spell these words in the abbreviated 
fashion whenever they wanted their readers to pronounce 
them in that way, while prose writers, unconcerned about 
the pronunciation given to their words, retained the full 
forms in spelling. The next step was that the short forms 
were branded as vulgar by schoolmasters with so great a 
success that they disappeared from ordinary conver- 
sation while they were still retained in poetry. And now 
they are distinctly poetic and as such above the reach of 
common mortals. 

233. Among the elements of ordinary language, some 
can be traced back to individual authors. Besides those 
already mentioned I shall cite only a few. Surround 
originally meant to overflow (Fr. sur-onder, Lat. super- 
undare); but according to Skeat, both the modern sig- 
nification, which implies an erroneous reference to round, 
and the currency of the word are due to Milton. The soft 
impeachment is one of Mrs. Malapropos expressions (in 
Sheridans's Rivals, act V, sc. 3). Henchman was made 
generally known by Scott, and to croon by Burns. Burke 
originated the expression 'the Great Unwashed.' A certain 
number of proper names in works of literature have been 
popular enough to pass into ordinary language as appela- 
tives^, as for instance pander or pandar from Chaucer's 
Troilus and Criseyde, Abigail 'a servant-girl' from Beau- 
mont and Fletcher's Scornful Lady, Mrs. Grundy as a 
personification of middle-class ideas of propriety from 
Morton's Speed the Plough, Paul Pry 'a meddlesome busy- 
body' from Poole's comedy of that name, Sarah Gamp 
'sick nurse of the old-fashioned type' and 'big umbrella' 
from Dickens's Martin Chuzzlewit, Pecksniff 'hypocrite' 


I Aronstein, Englische Studien XXV, p. 245 ff., Josef Rei- 
nius , On Transferred Appellations of Human Beings, Goteborg 
1903, p. 44ff 

232 IX. Shakespeare and the Language of Poetry. 

from the same novel, Sherlock Holmes 'acute detective' 
from Conan Doyle's stories. 

234. Ordinary language sometimes makes use of the 
same instruments as poetry. Above (§ 56) we have seen 
a number of alliterative formulas; here I shall give some 
instances of riming locutions : highways and byways, town 
and gown, it will neither make nor break me (cf. the 
alliterative make . . . mar), fairly and squarely, toiling 
and moiling, as snug as a bug in a rug (Kipling), rough 
and gruff, 'I mean to take that girl — snatch or catch' 
(Meredith), moans and groans^. Compare also such 
popular words as handy-dandy, hanky-panky, namby- 
pamby, hurly-burly, hurdy-gurdy, hugger-mugger, hocus 
pocus, hoity toity or highly tighty, higgledy-piggledy or 
higglety-pigglety, hickery- pickery. Hotchpot (from French 
hocher 'shake together' and pot) was made hotchpotch for 
the sake of the rime; then the final tch was changed into 
dge (cf. knowledge from knowleche) : *hotchpodge, and the 
rime was re-established: hodgepodge. 

235. Rhythm undoubtedly plays a great part in ord- 
'inary language, apart from poetry and artistic (or arti- 
ficial) prose. It may not always be easy to demonstrate 
this; but in combinations of a monosyllable and a di- 
syllabic by means of and the usual practice is to place 
the short word first, because the rhythm then becomes 
the regular 'aa 'aa instead of 'aaa 'a ('before the a denotes 
the strongly stressed syllable). Thus we say 'bread and 
butter', not 'butter and bread'; further: bread and water, 
milk und water, cup and saucer, wind and weather, head 
and shoulders, by fits and snatches, from top to bottom, 

I As Old English has mcenan 'moan', the modem verb may 
have derived its vowel from the frequent collocation with groan, 
OE. granian. Square may owe one of its significations to the 
collocation with fair. 

Rime and Rhythm. 233 

rough and ready, rough and tumble, free and easy, dark 
and dreary, high and mighty, up and doing^. It is pro- 
bable that rhythm has also played a great part in deter- 
mining the order of words in other fixed groups of greater 
complexity. 2 

1 Compare also such titles of books as Songs and Poems, 
Men and Women, Past and Present, French and English, Night 
and Morning. In some instances, rhythm is obviously not the 
only reason for the order, but in all I think it has been at 
least a concurrent cause. 

2 P. Fijn van Draat, Rhythm in English Prose (Heidelberg 
1 9 10) has many interesting observations on the influence of 
rhythm, though I would not subscribe to all his conclusions. 

Chapter X. 

236. In the preceding chapters we have considered the 
early vicissitudes of the EngUsh language, the various 
foreign influences brought from time to time to bear on 
it, its inner growth, lexical and grammatical, and the 
linguistic tendencies of its poets. It now remains to look 
at a few things which have contributed towards shaping 
the language, but which could find no convenient place 
in any of the preceding chapters, and then to say some- 
thing about the spread and probable future of the lan- 

237. Aristocratic and democratic tendencies in a nation 
often show themselves in its speech; indeed, we have 
already regarded the adoption of French and Latin words 
from that point of view. It is often said, on the Continent 
at least, that the typical Englishman's self-assertion is 
shown by the fact that his is the only language in which 
the pronoun of the first person is written with a capital 
letter, while in some other languages it is the second 
person that is honoured by this distinction, especially 
the pronoun of courtesy (Germain Sie, often also Du, 
Danish De and in former times Du, Italian Ella^ Lei, 
Spanish V. or Vd., Finnish Te). Weise goes so far as to 
say that 'the Englishman, who as the ruler of the seas 
looks down in contempt on the rest of Europe, writes 
in his language nothing but the beloved / with a big 

Aristocratic? 235 

letter'.^ But this is little short of calumny. If self- 
assertion had been the real cause, why should not me 
also be written Me} The reason for writing / is a much 
more innocent one, namely the orthographic habit in the 
middle ages of using a 'long i' (that is, j or I), whenever 
the letter was isolated or formed the last letter of a group; 
the numeral one was written j or I (and three, iij, etc.) 
just as much as the pronoun. Thus no sociological in- 
ference can be drawn from this peculiarity. 

238. On the other hand, the habit of addressing a single 
person by means of a plural pronoun was decidedly in its 
origin an outcome of an aristocratic tendency towards 
class-distinction. The habit originated with the Roman 
Emperors, who desired to be addressed as beings worth 
more than a single ordinary man; and French courtesy in 
the middle ages propagated it throughout Europe. In 
England as elsewhere this plural pronoun (you, ye) was 
long confined to respectful address. Superior persons or 
strangers were addressed as you; thou thus becoming the 
mark either of the inferiority of the person spoken to, 
or of familiarity or even intimacy or affection between 
the two interlocutors. English is the only language that 
has got rid of this useless distinction. The Quakers (the 
Society of Friends) objected to the habit as obscuring 
the equality of all human beings; they therefore thou'd 
(or rather thee'd) everybody. But the same democratic 
levelling that they wanted to effect in this way, was 
achieved a century and a half later in society at large, 
though in a roundabout manner, when the pronoun you 
was gradually extended to lower classes and thus lost 
more and more of its previous character of deference. 
Thou then for some time was reserved for religious and 
literary use as well as for foul abuse, until finally the 

I Charakteristik der lateinischen Sprache. 1899, p. 21. 

236 X- Conclusion. 

latter use was discontinued also and you became the only- 
form used in ordinary conversation. 

239. Apart from the not very significant survival of 
thou, English has thus attained the only manner of ad- 
dress worthy of a nation that respects the elementary 
rights of each individual. People who express regret at 
not having a pronoun of endearment and who insist how 
pretty it is in other languages when, for instance, two 
lovers pass from vous to the more familiar tu, should con- 
sider that no foreign language has really a pronoun ex- 
clusively for the most intimate relations. Where the two 
forms of address do survive, thou is very often, most often 
perhaps, used without real affection, nay very frequently 
in contempt or frank abuse. Besides, it is often painful 
to have to choose between the two forms, as people may 
be offended, sometimes by the too familiar, and some- 
times by the too distant mode. Some of the unpleasant 
feeling of Helmer towards Krogstad in Ibsen's Dukkehjem 
('A Doll's House' or 'Nora') must be lost to an English 
audience because occasioned by the latter using an old 
schoolfellow's privilege of thou-ing Helmer. In some 
languages the pronoun of respect often is a cause of 
ambiguity, in German and Danish by the identity in 
form of Sie (De) with the plural of the third person, in 
Italian and Portuguese by the identity with the singular 
(feminine) of the third person. When all the artificialities 
of the modes of address in different nations are taken 
into account — the Lei, Ella, voi and tu of the Italians, 
the vossa merce ('your grace', to shopkeepers) and voce 
(shortened form of the same, to people of a lower grade) 
of the Portuguese (who in addressing equals or superiors 
use the third person singular of the verb without any 
pronoun or noun), the gij, jij, je and U of the Dutch, 
not to mention the eternal use of titles as pronouns in 
German and, still more, in Swedish ('What does Mr. 

You. The Bible. 237 

Doctor want?' 'The gracious Miss is probably aware', 
€tc,) — the English may be justly proud of having avoid- 
ed all such mannerisms and ridiculous extravagances, 
though the simple Old English way of using thou in ad- 
dressing one person and ye in addressing more than one 
would have been still better. 

240. Religion has had no small influence on the Enghsh 
language. The Bible has been studied and quoted in 
England more than in any other Christian country, and 
a great many Bibhcal phrases have passed into the 
ordinary language as household words. The style of the 
Authorized Version has been greatly admired by many 
of the best judges of English style, who — with some 
exaggeration — recommend an early familiarity with 
and a constant study of the English bible (and of that 
great imitator of Biblical simplicity and earnestness, John 
Bunyan) as the best training in the English language.^ 
Tennyson found that parts of The Book of the Revelation 
were finer in English than in Greek, and he said that 

I See the long series of quotations given in Albert S. Cook's 
little book 'The Bible and English Prose Style' (Boston, 1892). 
On the other hand, Fitzedward Hall says, 'To Dr. Newman, 
and to the myriads who think as he does about our English 
Bible, one would be allowed to whisper, that the poor 'Turks' 
of the Prayer Book talk exactly in their own fashion, and for 
reasons strictly analogous to theirs, about the purity of diction, 
and what not, of 'the Blessed Koran' .... Ever since the 
Reformation, the ruling language of English religion has been, 
with rare exception, an affair either of studied antiquarianism 
or of nauseous pedantry. Simphcity, and little more, was 
aimed at, originally; and it sufficed for times of real earnest- 
ness. But the very quaintness of phrase which King James 
countersigned has attained to be canonized, till a hath, or 
a thou, delivered with conventional unction , now well nigh 
inspires a sensation of solemnity in its hearer, and a per- 
suasion of the sanctanimity of its utterer'. {Modern English 
p. 16—17.) 

238 X. Conclusion. 

'the Bible ought to be read, were it only for the sake of 
the grand English in which it is written, an education in 
itself.'^ The rhythmical character of the Authorized 
Version is seen, for instance, in the well-known passage 
(Job III. 17) 'There the wicked cease from troubling: 
and there the wearie be at rest', which Tennyson was able 
to use as the last line of his 'May Queen' with scarcely 
any alteration: 'And the wicked cease from troubling, 
and the weary are at rest'. 

241. C. Stoffel has collected quite a number of scriptural 
phrases and allusions used in Modern English^, such as 
'Tell it not in Gath', 'the powers that be', 'olive branches' 
(children), 'strain at [or out) a gnat', 'to spoil the Egyp- 
tians', 'he may run that readeth it', 'take up his parable', 
'wash one's hands of something, 'a still small voice', 
'thy speech bewrayeth thee'. Some which Stoffel does 
not mention may find their place here. The modern word 
a helpmate is a corruption of the two words in Gen. II, 18: 
'I will make him an helpe meet for him' [meet 'suitable'); 
the slang word a rib 'a wife' is from Genesis, too, and so 
is the expression 'the lesser lights'. 'A howling wilderness' 
is from Deuteron. XXXII. 10. 'My heart was still hot 
within me; then spake I with my tongue' (used, for in- 
stance, in Charlotte Bronte's 'The Professor', p. 161) is 
from Psalms XXXIX. 3, and 'many inventions' from 
Ecclesiastes VII. 29. From the New Testament may be 
mentioned 'to kill the fatted calf'^, 'whited sepulchres', 
'of the earth, earthy', and 'to comprehend with all 
saints , what is the breadth , and length , and depth and 

1 Life and Letters, II. 41 and 71. 

2 Studies in English, Written and Spoken, 1894, p. 125. 

3 While the phrase prodigal son is not found in the 
text of the Bible, it occurs in the heading of the chapter 
(Luke XV). 

Scriptural Words. 2^0 

242. The scriptural 'holy of holies', which contains a 
Hebrew manner of expressing the superlative^ has given 
rise to a great many similar phrases in English, such as 
'in my heart of hearts' (Shakesp. Hamlet, HI. 2. 78; 
Wordsw. Prelude XIV. 281), 'the place of all places' 
(Miss Austen, Mansf. P. 71), 'I rememberj'you a buck of 
bucks' (Thackeray, Newc. lOO), 'every lad has a friend 
of friends, a crony of cronies, whom he cherishes in his 
heart of hearts' (ib. 148), 'the evil of evils in our present 
politics' (Lecky, Democr. and Lib. I. 21), 'the woman is 
a horror of horrors' (H. James, Two Magics 60), 'that 
mystery of mysteries, the beginning of things' (Sully, 
Study of Childh. 71), 'she is a modern of the moderns' 
(Mrs. H. Ward, Eleanor 265), 'love like yours is the pearl 
of pearls, and he who wins it is prince of princes' (Hall 
Caine, Christian 443), 'chemistry had been the study of 
studies for T. Sandys' (Barrie, Tommy and Grizel 6). 
Compare also 'I am sorrowful to my tail's tail' (Kipling, 
Sec. Jungle B. 160). 

243. Some scriptural proper names have often been 
used as appellatives, such as Jezebel and Rahab; when a 
driver is called a jehu in slang, the allusion is to 2 Kings 
IX. 20, where Jehu's furious driving is mentioned^. There 
is an American slang expression 'to give a person Jessie' 
meaning, 'to beat him soundly', which is not explained 
in the Dictionaries (quotations may be found in Bartlett 
and in Farmer and Henley). Is it not in allusion to the 
rod mentioned in Isa. II. 1} (There shall come forth a 
rod out of the stem of Jesse.') The N. E. D. has the 
spelling Jesse with the meaning 'a genealogical tree re- 

1 Cf. I Timothy VI. 15 'the King of kings, and Lord of 

2 Y or j or am or jorum 'drinking bowl' 2lT\A jerry see N.E. D., 
where 2 Sam. VIII. 10, and Stoffel, Studies in Engl. 138, where 
I King XIV. 10 is quoted. 


2AO X. Conclusion. 

presenting the genealogy of Christ ... a decoration for a 
wall, window, vestment, etc., or in the form of a large 
branched candlestick'. 

244. The influence of Puritans, though not strong 
enough to proscribe such words as Christmas, for which 
they wanted to substitute Christtide in order to avoid 
the Catholic mass, was yet strong enough to modify the 
custom of swearing. In Catholic times all sorts of fan- 
tastic oaths were fashionable: 

Hir othes been so grate and so dampnable, 

That it is grisly for to here hem swere; 

Our blissed lordes body they to-tere; * 

Hem thoughte Jewes rente him noght ynough. ^ 

This practice was continued after the Reformation, and 
all sorts of alterations were made in the name of God in 
order to soften down the oaths: gog, cocke, gosse, gom, 
Gough, Gad, etc. Similarly instead of (the) Lord people 
would say something like Law, Lawks, Losh, etc. Some- 
times only the first sound was left out (Odd's lifelings, 
Shakesp. Tw. V. 187), more often only the genitive ending 
survived: 'Sblood (God's blood), 'snails, 'sHght, 'slid, 
'zounds (God's wounds). The final sound of the nomina- 
tive is kept in 'drot it (God rot it), which was later made 
drat it (or with a playful corruption rabbit it). Many of 
these disguised oaths were extremely popular, and some 
survive to this day. Goodness gracious me, which defies 
all grammatical analysis, is one among numerous 
compromises between the inclination to swear and 
the fear of swearing; note also Rosalind's words: 'By 
my troth, and in good earnest, and so God mend 
mee, and by all pretty oathes that are not dangerous'. 
(As IV. I. 192.) 

I Chaucer C. T., C. 472 fi., also see Skeat's note to this passage, 
Chaucer's Works V p. 275. 

Profane Language. 241 

245. The Puritans caused a law to be enacted in 1606 
by which profane language was prohibited on the stage 
(3 James I. chap. 21), and consequently words like 
'zounds were changed or omitted in Shakespearian plays, 
as we see from a comparison of the folio of 1623 and the 
earlier quartos; Heaven or Jove was substituted for God, 
and 'fore me (afore me) or trust me for (a) fore God; 'God 
give thee the spirit of persuasion' (H 4 A I. 2. 170) was 
changed into 'Maist thou have the spirit of perswasion', etc. 
But in ordinary life people went on swearing, and from 
the comedies of the Restoration period a rich harvest 
may be reaped af all sorts of curious oaths. By little and 
little, however, the Puritan spirit conquered, and now 
there can be little doubt that the English swear less than 
other European nations and that when they do swear the 
expressions are more innocent than elsewhere. Even the 
usual terms for oaths, — 'profane language' and 'ex- 
pletives' — point to a greater purity in this respect. Where 
a French or German or Scandinavian lady will express 
surprise or a little fright by exclaiming (My) God 1, an 
Englishwoman will say Dear me! or Oh my! or Good 
gracious! Note also euphemisms Hke 'deuce' for devil 
and 'the other place' or 'a very uncomfortable place' 
for hell^. Among tabooed words in English one finds a 
great number which in other countries would be con- 
sidered quite innocent, and the English have shown a 
really astonishing inventiveness in 'apologies' for strong 
words of every kind. Damn is now considered extremely 
objectionable, and even such a mild substitute for it as 
confound is scarcely allowed in polite society^. In Bernard 
Shaw's Candida Morell is provoked into exclaiming 'Con- 
found your impudence !', whereupon his vulgar father- 

1 Compare also 'I will see you further'. 

2 In the original sense it has often to be accompanied by 
togethef to avoid misunderstanding. 

Jespbrskn: English, and ed. 16 

2j^2 ^- Conclusion. 

in-law retorts, 'Is that becomin language for a clorgy- 
man?' and Morell replies, 'No, sir, it is not becoming 
language for a clergyman. I should have said damn your 
impudence: thats what St. Paul or any honest priest 
would have said to you'. Other substitutes for damned 
are hanged, somethinged (much rarer) ^ and a few that 
originate in the manner in which the objectionable word 
is — not printed : dashed (a — or 'dash' being put instead 
of it), blanked (from the same manner), deed (from the 
abbreviation d — d; sometimes the verb is printed to D). 
Darned must be explained as a purely phonetical develop- 
ment of damned, which is not without analogies, while 
danged, which occurs in Tennyson, is a curious blending 
of damned and hanged^. Thus we have here a whole family 
of words with an initial d, allowing the speaker to begin 
as if he were going to say the prohibited word, and then 
to turn off into more innocent channels. The same is the 
case with the ^/-words. Blessed by a process which is 
found in other similar cases^ came to mean the opposite 
of the original meaning and became a synonym of cursed; 
blamed had the same signification^ Instead of these 
strong expressions people began to use other adjectives, 
shunting off after pronouncing bl- into some innocent 
word like bloody, which soon became a great favourite 
with the vulgar and therefore a horror to ears polite, or 
blooming, which had the same unhappy fate in the 
latter half of the nineteenth century. Few authors would 

1 Cf, the similar use of something in 'Where the something 
are you coming to?' (Pett Ridge, Lost Property 167.) 

2 'I'm doomed!' Corp muttered to himself, pronouncing 
it in another way. (Barrie, Tommy and Grizel, p. 122.) This 
shows another way of disguising the word in print. 

3 Cf. silly., French benet, etc. 

4 There exists also a word blamed , a blending of blamed 
and damned (darned). 


Objectionable Words, 243 

now venture to term their heroines 'blooming young girls' 
as George Eliot does repeatedly in 'Middlemarch'. Simil- 
arly Shakespeare's expression 'the bloody book of law' 
is completely spoilt to modern readers, and lexicographers 
now have to render Old English blodig and the correspond- 
ing words in foreign languages by 'bleeding', 'blood- 
stained', 'sanguinary' or 'ensanguined'; but even san- 
guinary is often made a substitute for 'bloody' in report- 
ing vulgar speech. 

246. This is the usual destiny of euphemisms; in order 
to avoid the real name of what is thought indecent or 
improper people use some innocent word. But when that 
becomes habitual in this sense it becomes just as ob- 
jectionable as the word it has ousted and now is rejected 
in its turn. Privy is the regular English development of 
French prive; but when it came to be used as a noun for 
*a privy place' and in the phrase 'the privy parts', it had 
to be supplanted in the original sense by private, except 
in 'Privy Council', 'Privy Seal' and 'Privy Purse', where 
its official dignity kept it alive. The plural parts was 
an ordinary expression for 'talents, mental ability', 
until the use of the word in veiled language made it 

247. I do not know whether American and especially 
Boston ladies are really as prudish as they are reported 
to be, speaking of the limhs of a piano and of their own 
benders instead of legs, or saying waist instead of hody^. 

1 Cf. from America 'He -biddy. — A male fowl. A product 
of prudery and squeamishness'. Farmer, Americanisms -^.'z^l- 
Cf. also Storm, Engl. Philologie , p. 887 (roosterswain). 

2 See Thackeray, Virginians y quoted by Hoppe, Supple- 
mentlexicon, s. v. leg; Bartlett's and Farmer's Dictionaries of 
Americanisms , etc. Cf. also Opie Read , A Kentucky Colofiel, 
p. II 'He was so delicate of expression that he always said 
limb when he meant leg'. 


244 ^' Conclusion. 

But when to alter is said in the Southern States instead 
of to geld, and when ox is commonly used in America 
for bull (jocosely even gentleman cow!)'^, the same tend- 
ency may be observed on this side the Atlantic too. At 
least Mr. F. T. Elworthy, who knows the ways of Somerset 
peasants better than anybody else, says that the plain 
old English names for the male animals are going out of 
use: 'It has, perhaps, been taught or implied that such 
names as Bull, Stalhon, Boar, Cock, Ram are in- 
delicate; at any rate, we must no longer call a spade a 
spade, but there is a very distinct tendency to fine them 
down, by a weakening process, so that at last the generic 
word for the animal has commonly got to be used to 
express the entire male' (Elworthy, Fresh Words and 
Phrases in the Somersetshire Dialect, p. 6^). I am afraid 
we have here alighted on a trait which does not bear out 
my description (in the introductory chapter) of English 
as a masculine language. However, it is possible that the 
tendency here mentioned may be a passing one only and 
that common sense will prevail — as it has prevailed in 
the case of trousers, which word is now certainly less 
proscribed than it was fifty years ago. Perhaps the very 
absurdity of the taboo, which made people invent no 
end of comic names (inexpressibles, inexplicables, in- 
describables, ineffables, unmentionables, unwhisperables, 
my mustn't-mention-em, sit-upons, sine qua nons, etc.) 
has been the reason of the re-instatement of the good old 
word. Prudery is an exaggeration, but purity is a virtue, 
and there can be no doubt that the speech of the average 

1 'One sometimes sees a 'lady -dog' offered for sale in 
England, but 'male -sheep', 'male -hogs', 'gentlemen -turkeys', 
and 'gentlemen -game -chickens' belong to the natural history 
of refined Boston only.' T. Baron Russell, Current Ameri- 
canisms 1 6. 

2 Transactions of the Philological Society, 1898. 

Prudery. 245 

Englishman is less tainted with indecencies of various 
kinds than that of the average continental. — 

248. This volume has in so far been one-sided as it 
has dealt chiefly with Standard English and has left out 
of account nearly everything that is not generally ac- 
cepted as such, apart from here and there a nonce-for- 
mation or a bold expression which is not recognized as 
good English though interesting as showing the possibili- 
ties of the language and perhaps in some cases deserving 
popularity just as well as many things that nobody finds 
fault with. The question how one form of English came 
to be taken as standard in preference to dialects, has been 
deliberately omitted as well as all the problems connected 
with that pseudo-historical and anti-educational abom- 
ination, the English speUing.^ What I have to say 
on these subjects and on provincialisms, cockneyisms and 
vulgarisms, cant, slang, American and Colonial English, 
Pidgin- English and Negro- English, etc., must be left for 
the future; at present I shall conclude with a'few remarks 
on what might be called the Expansion of English. 

249. Only two or three centuries ago, English was 
spoken by so few people that no one could dream of its 
ever becoming a world language. In 1582 Richard Mul- 
caster wrote. The English tongue is of small reach, stretch- 
ing no further than this island of ours, nay not there 
over air. 'In one of Florio's Anglo- Italian dialogues, an 
Italian in England, asked to give his opinion of the 
language, replied that it was worthless beyond Dover. 
Ancillon regretted that the English authors chose to 
write in English as no one abroad could read them. Even 
such as learned English by necessity speedily forgot it. 

I A historical account of the English sound -system and 
English spelling may be found in my Modem English Gram- 
mar I (Heidelberg, Carl Winter, 1909). 


2 /1 6 ^- Conclusion. 

As late as 1718, Le Clerc deplored the small number of 
scholars on the Continent able to read English'.^ Compare 
what Portia replies to Nerissa's question about Faucon- 
bridge, the young baron of England (Merch. I. 2. 72) : 
'You know I say nothing to him, for hee understands not 
me, nor I him: he hath neither Latine, French, nor Italian, 
and you will come into the Court and sweare that I have 
a poore pennie-worth in the English. Hee is a proper 
mans picture, but alas, who can converse with a dumbe 
show?' In 17 14 Veneroni published an Imperial Diction- 
ary of the four chief languages of Europe, that is, Italian, 
French, German and Latin^. Now, no one would over- 
look English in making even the shortest possible list of 
the chief languages, because in political, social, and 
literary importance it is second to none and because it is 
the mother-tongue of a greater number of human beings 
than any of its competitors. 

250. It would be unreasonable to suppose, as is some- 
times done, that the cause of the enormous propagation 
of the English language is to be sought in its intrinsic 
merits. When two languages compete, the victory does 
not fall to the most perfect language as such. Nor is it 
always the nation whose culture is superior that makes 
the nation of inferior culture adopt its language: in some 
parts of Switzerland German is gaining ground at the 
expense of French, and in others French is supplanting 
German, yet no one can suppose that the superiority of 
the two nations is reversed in two adjacent districts. It 
sometimes happens in a district of mixed nationalities 

1 Ch. Bastide, Huguenot Thought in England. Journal of \ 
Comparative Literature I (1903) p. 45. 

2 Das kayserliche Spruch- und Worterbuch, darinnen die 
4 europaischen Hauptsprachen , als nemlicti: das Italianische, 
das Frantzosische, das Teutsche und das Lateinische erklart 

Expansion of English. 247 

that the population which is intellectually superior give 
up their own language because they can learn their 
neighbours' tongue while these are too dull to learn 
anything but their own: this is said by some to be the 
reason why in Posen and adjacent districts Polish is 
gaining ground over German, a fact which others ascribe 
to the greater fertility of the Poles. A great many social 
problems are involved in the general question of rivalry 
of languages^, and it would be an interesting, but difficult 
task to examine in detail all the different reasons that have 
in so many regions of the world determined the victory 
of English over other languages , European and non- 
European. Political ascendancy would probably be found 
in most cases to have been the most powerful influence. 

251. However that may be, the fact remains that no 
other European language has spread over such vast 
regions during the last few centuries, as shown by the 
following figures, which represent the number of millions 
of people speaking each of the languages enumerated^: 

Year English German Russian French Spanish Italian 




































1 Some excellent remarks may be found in H. Morf, Deutsche 
und Romanen in der Schweiz (Ziirich 1901). See also Will's 
dissertation, quoted below. 

2 See Lewis Carnac, quoted by R. M. Meyer, Indogermani- 
sche Forschunge7i XII, 84; E. Hasse, Handworterbuch der 
Slaatswtssenschafteji, 'Kolonien und Kolonialpolitik'; Otto Will, 
Die Tauglichkeit und die Aussichten der englischeii Sprache ah 
Weltsprache , Breslau 1903 The numbers given are necessarily 
approximative only, especially for the older periods. Where 
my authorities disagree, I have given the lowest and in paren- 
thesis the highest figure. 


X. Conclusion. 

.Whatever a remote future may have in store, one 
need not be a great prophet to predict that in the near 
future the number of Enghsh-speaking people will in- 
crease considerably. The curse of Babel is beginning to 
lose its sting, and it must be a source of gratification to 
mankind that the tongue spoken by two of the greatest 
powers of the world is so noble, so rich, so pliant, so ex- 
pressive, and so interesting as the language whose growth 
and structure I have been here endeavouring to charac- 

Phonetic Symbols. 

(Alphabet of the Association Phonitique Internationale.) 

' stands before the stressed syllable. 

• indicates lenght of the preceding vowel. 

[a-] as in alms. 

[ai] as in /ce. 

[au] as in h.oust. 

[ae] as in h.aX.. 

[ei] as in h<2te. 

[9] as in about, colour. 

[i-] as in French d/se. 

[ij] as in h£aX\ practically 

= [i •]. 

[ou] as in s^. 

[o] as in h<9t. 

[o] as in hall. 

See my Modern English Gra?nmar (1909). 

[a] as in h«t. 

[u-] as in French dpowse. 

[uw] as in who; practically 

= [u-]. 
[y] as in French vu. 

[]?] as in thiii. 

[d] as in th\s. 

[s] as in jeal. 

[z] as in zedX. 

[f] as in shin\ [tj] as in ch\n. 

[3] as in vij/on; [dsjasin^n. 


O. E. = Old English ('Anglo-Saxon'). 

M.E. = Middle English. 
Mod. E. = Modem English. 

O. Fr. = Old French. 

O. N. = Old Norse. 
O. H. G. = Old High German. 

N. E. D. = A New English Dictionary, by Murray, Bradley, 
and Craigie. 

The titles of Shakespeare's plays are abbreviated as in 
Al. Schmidt's Shakespeare- Lexicon, thus Ado = Much Ado about 
Nothing, Gent. = The two Gentlemen of Verona, H4A = First 
Part of Henry the Fourth, Hml. = Ha?nlet, R2 = Richard the 
Second, Tp. = Tempest, Tw. = Twelfth Night, Wiv. = The 
Merry Wives of Wi?tdso? , etc. Acts , scenes , and lines as in 
the Globe edition. 


References are to the number of the sections. 
Only the more important words used as examples are included. 

a pronoun 72. 

abbreviations 10, 176. 

Abigail 233. 

-able, 108, 109. 

absolute participle 125. 

abstract terms 114 if. 

academies 18. 

accent, see stress and tone. 

accidence 178 ff. 

-aceous 123. 

ache 169. 

accommodate 219. 

action, deed, loi. 

activity 219. 

Addison, on who and which 

adjectives, place 85, — Latin 

and English i3iff., in -ish 

adventure 116. 
adverbs turned into adjectives 

advice 116. 

Africa, Dutch and English 1 54. 
agent-nouns 162. 
aggravate 119. 
aggressive no. 
aid, help 100. 
aim III, 
Alfred 46, 48, 53, 58, 59. 

alliteration 54, 56. 

alms 187. 

also in Shakespeare and Bacon 

am (reading) 206. 
ambiguity 140, 172. 
America, speech-mixture 78, 

prudery 247. 
ana, 123, 
anchor 32. 
Ancrene Riwle, French words 

in, 94. 
angel 38, 86. 
Angles 34. 
Anglicizing of Scandinavian 

words 63. 
Anglo-Saxon, see Old English. 
anti- 124. 
April 116. 
aquiline 132. 
archaisms 229. 
Arian family of languages 21, 

character of primitive Arian 

-arious 123. 
Aristarchy 143 note, 
aristocratic tendencies 82ft"., 

93, 130, 237. 
art, words relating to, qi. 
article, definite 9. 



Aryan, see Arian. 
assassination 219. 
■ation 123. 
-ative 123. 
Australasia 157. 
authors, expressions due to in- 
dividual authors 233. 
avunculaf 132. 
awe 70. 
ay 70. 

back-formations 173, 188, 189. 

Baconian theory 220 (p. 217). 

bairn 64, 

bankrupt 116. 

Banting 173. 

bath, bathe 168. 

bathos 119. 

beet 32. 

beg 173. 

Bell's phonetic nomenclature 

Beowulf 49, 54. 
Bible, influence 196, 231, 240 ff. 
birth 70, 
bit, bite 170. 
blend 64. 
blessed 245. 
bloody 131, 245. 
bloom -ji. 
blooming 245. 
bonnet 219. 
^<?c'/'/% 61. 
bound 61, 104. 
bourne 228. 
bread 71. 
breeches 191. 
^r^<?^ 170. 
brethren 191. 
bridal 210. 
Britons, see Celts. 
brood ijo. 
brother 191, 224. 

^rwj-A 171. 
^w^'/^ 61. 
(5«r^/tf 173. 
^«/^^ 173. 
-fty 60, 61, 74. 
by-law 74. 

^fl^ 176. 

^a^' 173. z'" 

call 59, 

^(xr/" 36. 

Caxton 69, 98. 

Celts 21, in England 35, Celtic 

words in English 36 ff. 
censuj'e 219. 
certainty, certitude 116. 
ch 112. 

^^(?a/ 32, 210. 
Charlie 112. 
charm 219. 
Chaucer 94, 98, 226. 
<;^<?^r 112. 
chick 56. 
children 191. 
children's words 177. 
choose, choice 97. 
Christianity, influence on 

language 38fif. 
church 38. 
classical studies, effect on style 

127; see also Latin and Greek. 
cleave 231. 
climax 119. 
clippings of long words 10, 

176 (173). 

clothe, dress 100. 

CO- 124. 

coined words I58f. 

cold, synonyms 136. 

colour and derivatives 117. 

companion 219. 

compounds, instead of adject- 
ives 132, verbs 174, nouns 210. 



conciseness lo. 

confound 245. 

conjunctions 209, 

consonants 3, groups 5, 6, 

shift 24, 26, in nouns and 

verbs 168 f. 
continuous forms 206. 
cook 32. 

cordial, hearty 100. 
cose 173. 
cottage, hut 100, 
cowl 39. 
crave 74. 

critic, critique, criticize 116. 
croon 233. 
cuisine 88. 
curse 36. 
, Cynewulf's First Riddle 58. 

dainty 210. 

dale 64, loi. 

dalliance no. 

^(2w;^ and substitutes 245. 

Danes, Danelaw 58, 61, cf. 

danger 112. 

D'Arblay, Madame 145. 
dafkle 173. 
dart 171. 

Darwin, on classical studies 127. 
de- 124. 
^^^/ 116. 

democratic tendencies 237. 
describe 116. 
devil 38. 
dialects, differences in verbal 

inflection 193. 
Dickens on a large retinue of 

words 135. 
die 61, 72. 
differentiations 66, 84, lOO, 

III, 112, 116, 167, 179. 
difficult 173. 

diminutives 13. 

disciple 39. 

dish 32. 

do 206, 225, 226; doeth, doth 

^^z/(J/ 112, 116. 
drat it 24^. 
dream 71. 

dress, words relating to, 90. 
dress, dressing 167. 
drown 51. 
Dryden, French words 95, 

syntax 126. 
duration no. 

Dutch in South Africa 154. 
duty III. 
dwell 71. 

e- and in- (im-) confounded 140. 

earl 71. 

Easte? 42. 

ecclesiastical terms, Latin 38 ff., 

French 86. 
-ed, suffix 162. 
edge 66. 
edify 133. 
-i?^ III. 
egg 66, 69. 
tf>^^ 220. 
'em 72. 
-f« nouns in, 160, verbs in 

162, plural of nouns 185, 

of verbs 193. 
endings, worn off, 7. 
English, masculinity of 2ff., a 

world language 2486. 
enormous 119. 
equal 116. 
•er 97, 162. 
etymology oipup, cad, pet i^ it 

unknown, of many short 

words 176. 
euphemisms 244 ff. 



«uphony 3!?., 226. 
Euphuism 218. 

.£X- 124. 

example^ exemplary 117. 
exhibit, exhibition 167. 
expansion of English 249 ff. 
•eye-words 142. 

/ alternating with v 168. 
Jad 176. 

faint 171. 

family, familiar 132. 

Jeed 170. 

feel, feeling 167. 

felicity 99. 

feminine nouns, formation of, 

feudalism 82. 

fierce 103. 
Jltz 103. 
flute 112. 

/^^ 56. 

/<?/>^, people 100. 

y27(?<ar 170. 

^^ with an infinitive 211. 

foreign titles 156. 

frame 171. 

iFrench 81 ff., rulers of England 
82, spheres of signification 
82 if., number of words in 
early authors 94, date of 
adoption 95, French and 
native words 97 f., not 
popularly understood 99, 
synonyms 100, forms 103, 
sounds 105, hybrids 106 ff., 
independant formations on 
English soil iioff, old and 
recent loans 112, French 
and Latin ii4ff 

^.friend 71. 

fro, from 66. 

future 8 1, 206. 

g, pronunciation 112. 

gain 76, 97. 

gait 76. 

games, terms of, 89. 

gate 76. 

gender 205. 

genitive case, Scandinavian 80, 

position 81, endings iSoflf. 
Germanic, pre-historic 20 ff., 

how considered by Romans 

23, invasion [of England 33 ff., 

in Romance countries 78. 
gerund 200 ff., see ing. 
gestic 143 note. 
get 70, 231, get clea? 224, 

I have got 224, gat 231. 
get-at-able 109. 
gift 70. 

Gill, on Latin influence 150. 
give 70. 

glass, glaze 168. 
God 42, compounds 45, in 

oaths 245. 
gospel 43, 45. 
gossip 171. 
gown 36. 
grammar, simplification of 80, 

160, 163, I78ff. 
greed 173. 
Greek 114'ff. 
Grimm's Law -Ji^ OA. 
group-genitive 180. 
grovel 173. 
Grundy, Mrs. 233. 

hale 66. 

hallow 42. 

handbook 47. 

haplology 186. 

har binge 173. 

harmony of language 141, 

harTy 97. 

have auxiliary 206. 



hawk 173. 

heathen 43. 

heaven 76 note. 

hegemony 142. 

helpmate 241. 

hence 68. 

henchman 233. 

henpeck 174. 

^d?r 72, 

heraldry 82. 

hodgepodge 234. 

/^<?/w 71. 

^^/j/, i-^/w/ 100. 

homiciae 133. 

housekeep 174. 

housel 42. 

humorous application of learn- 
ed words 122, 147. 

Huxley on the genius of Eng- 
lish and Latin 127. 

hybridity 41, 106, 107, 123. 

hyperbohcal expressions 11. 

/, the pronoun 237. 
-iacal 123. 

-ie 13- 

ifnpeachment 233. 

in-, causes ambiguity 140. 

inch 32. 

indispensable 109. 

Indo-European, see Arian. 

in/angthief 74. 

infinitives, French 104, syntax 

ing 106, 200 ff., as a noun 201, 
with an object 202, with 
adverbs 20 1 , tense and voice 
203, with a subject 204. 

inhabitable 140. 

insomnia, sleeplessness 138. 

intensity no. 

inter- 124. 

international 124. 

international words 138. 

intonation 12. 

inverted word-order 14. 

invoice 103 note. 

-ish, in verbs 104, in adjectives 

island, isle 97. 
-ism, -ist 122 {. 
Italian loan words 31, 151. 
-He 123. 
item 119. 
its 205. 
-ize 123. 

jackass in Australia 157 note. 

James 103. 

jaunty 112. 

jaw 171. 

jehu 243. 

Jesse 243. 

Jezebel 243. 

jocular classicisms 122, 147. 

Johnson, Dr. Samuel 126, 135, 

Jutes 34. 

-kin 13. 

kindergarten 153. 

kine 191. 

kingly, royal, regal 131. 

kirk 67. 

^m 170. 

kitchen 32. 

^«z/^ 75. 

Knut 60, 61. 

kodak 158. 

labour 56. 

labyrinth, adjectives from 132. 

lake 97. 

language 116, 

Latin, earliest loan-words 32, 
spoken in England 35, influ- 
ence in modern times ii4flf., 
French and Latin 115 ff., 



number of words 118, de- 
viations from Latin usage 
iigff., hybrids 123, style and 
syntax 125 ff,, benefits and 
disadvantages 128 ff. 

laugh, laughter 167. 

laughable 109. 

law 74. 

laze 173. 
K Layamon, French words in 94. 
p learned words I2i, 131, 132, 
138, 144, plurals 141. 

legal words Scandinavian 74, 
French 84 f. 

I -less 66. 
-let 13. 
levy 104. 
like 209. 
-ling 173. 

loan-words in general 30 f., 37, 

^ i54ff., technical 31, 32, 38 ff., 

73fif., 82ff, 121, I5iff., non- 

» technical 76 ff., 92 ff., 128 ff. 
logic in grammar 15. 
long words, psychological effect 

of 137- 
loose 66. 
loot 151. 
Lowell, on newspaper writing 


machine and derivatives 117. 
mag-nitude 133. 
P main 97. 

Malapropisms 143. 

male animals 247. 

manly and synonyms 133. 

manslaughter 133. 

many 97. 

matin, momi?tg 100. 

meaning of Shakespearian 

words 219 f. 
means 188. 

men and women, linguistically 

different, 7, 11, 12, 18. 
Micawber's style 135. 
mile 32. 
military words, Scandinavian 

73, French 83, others 151. 
mill 32. 
Milton, syntax 126, vocabulary 

214, 216, surround 233. 
mine 179. 
mint 32. 
Miss 175. 

mixed languages 37, 78. 
mob 176. 

monosyllabism, force of 8, 9. 
monosyllables from various 

sources 175 ff. 
monger 32. 
mortar 32. 

move, movement, motion 1O7. 
murder 133, 

musical terms, Italian 31. 
mutation, plurals 186, verbs 1 70. 
mutin, derivatives iii. 

National character i, 2, 5. 1*^) 
II, 12, 13, 14, 16, 17, 18, 28, 
50, 73, 92, 93. 148, 155, 237 ff, 
240 f., 244 ff. 

native words as contrasted with 
loan-words 41 ff. 

navy 176. 

nay 66. 

nephew 97. 

neuter, Scandinavian 79, Eng- 
lish 205. 

new words from unknown 
sources 177. 

no 66. 

nominative. Old French 103. 

Norman, see French. 

Norse, see Scandinavian. 

Nurwegians6i, cf.Scandinavian. 



notoriotis 219. 

nouns in -ef 162, and verbs 

163 ff., from verbs 166, 

becoming adjectives 210. 
now-a-days 220. 
number, concord 16, formation 

of plural 141, 185 ff. 
number of words I28ff., in 

individual vocabularies 2 1 4 ff. 
numerals 199. 

oaths 244! 

obscuration of vowels 26, 139. 

occupy 219. 

-ocracy 123. 

odd 76. 

of 181, 183, oj his 184, holy 
of holies 242. 

offer 39. 

Old English (Anglo-Saxon), 
relations to other Germanic 
languages 34, dialects 34, 53, 
loans from Celtic 36, influence 
of Christianity 38 ff., loans 
from Latin and Greek 38ff., 
native formations 41 ff., 
literary capacities 48, poetry 
49 ff., synonyms 49, seafaring 
terms 49, 50, prose 48, 55. 

-ology 123. 

once 209. 

one 208. 

Orrmulum, French words 94. 

participle, absolute 125, cf. ing 
and passive. 

pander 233. 

parts 246. 

passive, English 17, Scandi- 
navian 79, of ing 203, is being 
built 206. 

Paul Pry 233. 

pea, pease 32, 188. 

pear 32. 

PecksnifJ^ 233. 

pedantry, absence of, 16, 17. 

peadle i']'^. 

peppe? 32. 

perfect 116. 

perfect 206. , 

periphrastic tenses 15, 206. 

pet 173. 

petty 84. 

phrases used attributively 17, 

French 92. 
phthisis 142. 
picture 116. 
place-names, Scandinavian 60, 

translated 156. 
plough 71. 
plunder 151. 
plural, learned formations 141, 

ordinary 185 ff., raised to a 

second power 191, unchanged 

192, of verbs 193. 
poetry. Old English 49, its form 

54, language of poetry distinct 

from prose language 53, 

225 ff. 
political words, French 82. 
politician 21 g. 
ponder 119. 
pony 36. 
pre- 124. 
premises 119. 
prepositions, Latin and Greek 

124, place 126. 
privy 246. 
pro- 124, 
profane language, Act against 

progress in word-formation 

1 60, in grammar 178 ff. 
progressive tenses 15, 206. 
pronouns, Scandinavian 72, 76, 

English 126, 205, 208, 237 ff. 



pronunciation of learned words 

proper names, adjectives from, 

131, 139. 
prose, Old English 48, 55, cf. 

provoke 119. 
prudery 245 ff. 

pseudo-antique formations 230, 
punctilium 122, 
puisne, puny 84. 
pup 173. 
Puritanism 244 fF, 

gua?'i 112. 

I quasi-classical words 121, 122. 
quince 103 note. 
raise 66. 
re- 124. 
rear 66. 
reduplicated perfects 27. 
relative pronoun, omission 81, 
126, who, which, that 126, 
which 205. 
reliable 109. 
remodelling of French words 

113, 116. 
remorse 219. 
Renaissance 114. 
resolution, resolve 167. 
retort 165. 
rhinoceros 141. 
rhythm 235. 
rich 97. 
riches 187. 
richness of the English language 


riding 74. 

rimes, male and female 8. 

riming locutions 234. 

Robert of Gloucester 96. 

rout, route 112. 

rove 173. 

Jesphrsen : English. 2nd ed. 

S in French nominatives 103, 
voiceless in nouns, voiced 
in verbs 168, in genitives 
180 fF., in plurals 185 ff., 
s for ses 186; in verbs 
193 ff. 

sail 171. 

salon, saloon 112. 

sa7ne 72. 

Sarah Gamp 233. 

Saxons 34. 

Scandinavian 57fif., similarity 
with English 62, Anglicizing 
63, parallel forms 65 ff., 
influence on meaning 71, 
Scandinavian words readily 
associated with native words 
72, spheres of signification 
73ff., military words 73, 
legal terms 74, commonplace 
words 76, Scandinavian in 
U. S. 78, forms of loan-words 

79, influence on grammar 

80, 81. 

scientific nomenclature 114, 
121, 138. 

scie7itist 121. 

scriptural phrases 241. 

seat 71, 170. 

self 208. 

sell 170. 

sensible no. 

sentences, abbreviated 10, used 
attributively 17. 

sex and language 7, 11, 12, 18. 

Shakespeare 213 ff., range of 
vocabulary 214 ff., religious 
views 217, individual char- 
acters 218, Euphuism 218, 
meanings different from 
modern 219, Shylock 221, 
periods in Shakespeare's fife 
222, provincialisms 222, 



boldness of language 223, 
the First and Second Folios 
223, use of new words 224, 
poetic diction 225, words 
and phrases due to him 228. 

shall 81, 206. 

sheer 219. 

Sherlock Holmes 233. 

Shetland 78 (note p. 79). 

Shylock's language 221. 

sidle 173. 

simplification of grammar 80, 
160, 163, lySff. 

siste? 70. 

sit 170. 

size 133. 

sky 76 note. 

slang 176, 243, 244 ff. 

smoke 171. 

sobriety 11. 

sounds 3,26,1 39, sound-changes 
in French words 105, 112. 

specializing in primitive vocab- 
ularies 51 ff 

Spencer, Herbert, on classical 
studies 127, on long words 

Spenser, influence on poetic 

style 229. 
split infinitive 211. 
sport 89. 
squirearchy 123. 
stick, stitch 169. 
stress, French and English 

contrasted 28, in French 

words 105, in Latin and 

Greek 139, 
stress-shift, Germanic 25 — 28. 
strong verbs 29, 178. 
style, Old EngHsh 48, 49, 

Latin 127, use of synonyms 

98, 135, Johnsonese 144 ff., 

journalese 148. 

subjunctive 20b. 

succeed, success 219. 

suffixes i6off. 

surround 233. 

syllable construction 5. 

synonyms in Old English 49 ff, 
heaven, sky 76 note, collo- 
cated 98, 135, French and 
native 100, Latin and native 
133 ff., move, motion, feel, 
feeling, etc. 167. 

syntax 14, 15, 16, 17, Scandin- 
avian 80a, Latin 125!, geni- 
tive 180 ff., plural 187, i9of., 
ing 200 ff., verbs 206, 211, 
pronouns 208, conjunctions 
209, compounds 210, Shy- 
lock's 221, Shakespeare's 

take 79. 

telegraphic style 10. 
Tennyson, prefers Saxon words 

tense-system 15, 22, 29, 206. 
th voiceless in nouns, voiced 

in verbs 168, in third singular 

193 ff., in ordinals 199. 
that, omission 81, relative 

pronoun 126. 
thetiee 68. 

they, them, their 70, 72. 
thou 232, 237 f. 
thoughtread 174. 
thrall 74. 
tithe 42, 199. 
though 70. 
Thursday 70. 
till 64. 
tidings 63. 

to as a pro-infinitive 211. 
tone 12. 
town 36. 



trace 103 note, 
trades, names of, 91. 
tradespeople's coinages 158. 
transpire 119. 
trousers 247. 
trusteeship 1 1 1 . 
trustworthy 109. 
typewrite 174. 

unaccountable 109. 
undemocratic character of clas- 
sical words 143. 
uninhabitable 140. 
usance 221. 

value-stressing 26 ff., 105. 

venture 116. 

verbal noun 200 ff., see ing. 

verbs, strong 29, 178, weak 
29, form of French 104, in 
-en 162, relation to nouns 

163 ff. 

verdict 116. 

victuals ii6. 

vocabulary, fulness of, 18, 

I28ff, individual 214 ff. 
voiced and voiceless consonants 

in verbs and nouns 97, 168. 
vowel-differences between 

nouns and verbs 170. 

vowel -sounds obscured 

voyage 1 12. 


wag 219. 

want 72. 

wapentake 74. 

wash 52. 

weak verbs 29. 

weird 228. 

whence 68. 

which 126, 205, 208. 

tuho 208, for he who 125, 

Humble Petition of who and 

which 126. 
whole 66. 
will 81, 206. 
wi?tdow 75. 
wi?ie 32. 

w^y^, wireless 138, 171. 
women, language of, 7, 11, 

12, 18. 
word-formation I58ff., regular 

processes i6off. 
word-order 14, 207, adjectives 

after nouns 85. 
Wulfstan 48, 55. 

-/ 13. 
Yankee 188. 

you 179, 232, 237f. 
References are to sections, not to pages. 

Druck von B. G. Teubner in Dresden. 





This book is due on the last date stamped below, 
or on the date to which renewed. Renewals only: 

Tel. No. 642-3405 
Renewals may be made 4 days priod to date due. 
Renewed books are subject to immediate recall. 

atuu LD SEP 2 1 72 -8 AM 7 

Mflv 2 ^ 197 5 


REC. CIS. « 




K£C. CIR. M O'i '83 

OCT 2 8 19 83 ^ - 5 


BEC.C1R. J0LO5'83 

(N8837sl0)476— A-32 

General Library 
University of California ^ 



** >♦